“Sara
commited on
Commit
•
ac029b0
1
Parent(s):
5ba107d
adding more checkpoints
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- checkpoint-100/trainer_state.json +0 -179
- {checkpoint-100 → checkpoint-1000}/added_tokens.json +0 -0
- {checkpoint-100 → checkpoint-1000}/config.json +0 -0
- {checkpoint-100 → checkpoint-1000}/generation_config.json +0 -0
- {checkpoint-100 → checkpoint-1000}/model.safetensors.index.json +0 -0
- {checkpoint-100 → checkpoint-1000}/special_tokens_map.json +0 -0
- {checkpoint-100 → checkpoint-1000}/tokenizer.json +0 -0
- {checkpoint-100 → checkpoint-1000}/tokenizer.model +0 -0
- {checkpoint-100 → checkpoint-1000}/tokenizer_config.json +0 -0
- checkpoint-1000/trainer_state.json +1529 -0
- {checkpoint-100 → checkpoint-1000}/training_args.bin +0 -0
- {checkpoint-150 → checkpoint-1050}/added_tokens.json +0 -0
- {checkpoint-150 → checkpoint-1050}/config.json +0 -0
- {checkpoint-150 → checkpoint-1050}/generation_config.json +0 -0
- {checkpoint-150 → checkpoint-1050}/model.safetensors.index.json +0 -0
- {checkpoint-150 → checkpoint-1050}/special_tokens_map.json +0 -0
- {checkpoint-150 → checkpoint-1050}/tokenizer.json +0 -0
- {checkpoint-150 → checkpoint-1050}/tokenizer.model +0 -0
- {checkpoint-150 → checkpoint-1050}/tokenizer_config.json +0 -0
- checkpoint-1050/trainer_state.json +1604 -0
- {checkpoint-150 → checkpoint-1050}/training_args.bin +0 -0
- {checkpoint-200 → checkpoint-1100}/added_tokens.json +0 -0
- {checkpoint-200 → checkpoint-1100}/config.json +0 -0
- {checkpoint-200 → checkpoint-1100}/generation_config.json +0 -0
- {checkpoint-200 → checkpoint-1100}/model.safetensors.index.json +0 -0
- {checkpoint-200 → checkpoint-1100}/special_tokens_map.json +0 -0
- {checkpoint-200 → checkpoint-1100}/tokenizer.json +0 -0
- {checkpoint-200 → checkpoint-1100}/tokenizer.model +0 -0
- {checkpoint-200 → checkpoint-1100}/tokenizer_config.json +0 -0
- checkpoint-1100/trainer_state.json +1679 -0
- {checkpoint-200 → checkpoint-1100}/training_args.bin +0 -0
- {checkpoint-250 → checkpoint-1150}/added_tokens.json +0 -0
- {checkpoint-250 → checkpoint-1150}/config.json +0 -0
- {checkpoint-250 → checkpoint-1150}/generation_config.json +0 -0
- {checkpoint-250 → checkpoint-1150}/model.safetensors.index.json +0 -0
- {checkpoint-250 → checkpoint-1150}/special_tokens_map.json +0 -0
- {checkpoint-250 → checkpoint-1150}/tokenizer.json +0 -0
- {checkpoint-250 → checkpoint-1150}/tokenizer.model +0 -0
- {checkpoint-250 → checkpoint-1150}/tokenizer_config.json +0 -0
- checkpoint-1150/trainer_state.json +1754 -0
- {checkpoint-250 → checkpoint-1150}/training_args.bin +0 -0
- {checkpoint-300 → checkpoint-1200}/added_tokens.json +0 -0
- {checkpoint-300 → checkpoint-1200}/config.json +0 -0
- {checkpoint-300 → checkpoint-1200}/generation_config.json +0 -0
- {checkpoint-300 → checkpoint-1200}/model.safetensors.index.json +0 -0
- {checkpoint-300 → checkpoint-1200}/special_tokens_map.json +0 -0
- {checkpoint-300 → checkpoint-1200}/tokenizer.json +0 -0
- {checkpoint-300 → checkpoint-1200}/tokenizer.model +0 -0
- {checkpoint-300 → checkpoint-1200}/tokenizer_config.json +0 -0
- checkpoint-1200/trainer_state.json +1829 -0
checkpoint-100/trainer_state.json
DELETED
|
@@ -1,179 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": null,
|
| 3 |
-
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.3076923076923077,
|
| 5 |
-
"eval_steps": 10,
|
| 6 |
-
"global_step": 100,
|
| 7 |
-
"is_hyper_param_search": false,
|
| 8 |
-
"is_local_process_zero": true,
|
| 9 |
-
"is_world_process_zero": true,
|
| 10 |
-
"log_history": [
|
| 11 |
-
{
|
| 12 |
-
"epoch": 0.006153846153846154,
|
| 13 |
-
"eval_loss": 1.487821340560913,
|
| 14 |
-
"eval_runtime": 1.2374,
|
| 15 |
-
"eval_samples_per_second": 92.131,
|
| 16 |
-
"eval_steps_per_second": 4.849,
|
| 17 |
-
"step": 2
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.03076923076923077,
|
| 21 |
-
"grad_norm": 20.007999420166016,
|
| 22 |
-
"learning_rate": 6.153846153846155e-07,
|
| 23 |
-
"loss": 1.4703,
|
| 24 |
-
"step": 10
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.03076923076923077,
|
| 28 |
-
"eval_loss": 1.4221965074539185,
|
| 29 |
-
"eval_runtime": 1.1701,
|
| 30 |
-
"eval_samples_per_second": 97.431,
|
| 31 |
-
"eval_steps_per_second": 5.128,
|
| 32 |
-
"step": 10
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 0.06153846153846154,
|
| 36 |
-
"grad_norm": 29.657033920288086,
|
| 37 |
-
"learning_rate": 1.230769230769231e-06,
|
| 38 |
-
"loss": 1.4402,
|
| 39 |
-
"step": 20
|
| 40 |
-
},
|
| 41 |
-
{
|
| 42 |
-
"epoch": 0.06153846153846154,
|
| 43 |
-
"eval_loss": 1.238189935684204,
|
| 44 |
-
"eval_runtime": 1.1729,
|
| 45 |
-
"eval_samples_per_second": 97.195,
|
| 46 |
-
"eval_steps_per_second": 5.116,
|
| 47 |
-
"step": 20
|
| 48 |
-
},
|
| 49 |
-
{
|
| 50 |
-
"epoch": 0.09230769230769231,
|
| 51 |
-
"grad_norm": 10.80688190460205,
|
| 52 |
-
"learning_rate": 1.8461538461538465e-06,
|
| 53 |
-
"loss": 1.0957,
|
| 54 |
-
"step": 30
|
| 55 |
-
},
|
| 56 |
-
{
|
| 57 |
-
"epoch": 0.09230769230769231,
|
| 58 |
-
"eval_loss": 0.908875048160553,
|
| 59 |
-
"eval_runtime": 1.1739,
|
| 60 |
-
"eval_samples_per_second": 97.111,
|
| 61 |
-
"eval_steps_per_second": 5.111,
|
| 62 |
-
"step": 30
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 0.12307692307692308,
|
| 66 |
-
"grad_norm": 12.914667129516602,
|
| 67 |
-
"learning_rate": 2.461538461538462e-06,
|
| 68 |
-
"loss": 0.8136,
|
| 69 |
-
"step": 40
|
| 70 |
-
},
|
| 71 |
-
{
|
| 72 |
-
"epoch": 0.12307692307692308,
|
| 73 |
-
"eval_loss": 0.7436326146125793,
|
| 74 |
-
"eval_runtime": 1.1712,
|
| 75 |
-
"eval_samples_per_second": 97.34,
|
| 76 |
-
"eval_steps_per_second": 5.123,
|
| 77 |
-
"step": 40
|
| 78 |
-
},
|
| 79 |
-
{
|
| 80 |
-
"epoch": 0.15384615384615385,
|
| 81 |
-
"grad_norm": 5.206438064575195,
|
| 82 |
-
"learning_rate": 3.0769230769230774e-06,
|
| 83 |
-
"loss": 0.7864,
|
| 84 |
-
"step": 50
|
| 85 |
-
},
|
| 86 |
-
{
|
| 87 |
-
"epoch": 0.15384615384615385,
|
| 88 |
-
"eval_loss": 0.6924600601196289,
|
| 89 |
-
"eval_runtime": 1.1718,
|
| 90 |
-
"eval_samples_per_second": 97.287,
|
| 91 |
-
"eval_steps_per_second": 5.12,
|
| 92 |
-
"step": 50
|
| 93 |
-
},
|
| 94 |
-
{
|
| 95 |
-
"epoch": 0.18461538461538463,
|
| 96 |
-
"grad_norm": 5.670599460601807,
|
| 97 |
-
"learning_rate": 3.692307692307693e-06,
|
| 98 |
-
"loss": 0.622,
|
| 99 |
-
"step": 60
|
| 100 |
-
},
|
| 101 |
-
{
|
| 102 |
-
"epoch": 0.18461538461538463,
|
| 103 |
-
"eval_loss": 0.6791077852249146,
|
| 104 |
-
"eval_runtime": 1.1678,
|
| 105 |
-
"eval_samples_per_second": 97.62,
|
| 106 |
-
"eval_steps_per_second": 5.138,
|
| 107 |
-
"step": 60
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 0.2153846153846154,
|
| 111 |
-
"grad_norm": 3.6266868114471436,
|
| 112 |
-
"learning_rate": 4.307692307692308e-06,
|
| 113 |
-
"loss": 0.6672,
|
| 114 |
-
"step": 70
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 0.2153846153846154,
|
| 118 |
-
"eval_loss": 0.6631948947906494,
|
| 119 |
-
"eval_runtime": 1.1626,
|
| 120 |
-
"eval_samples_per_second": 98.06,
|
| 121 |
-
"eval_steps_per_second": 5.161,
|
| 122 |
-
"step": 70
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.24615384615384617,
|
| 126 |
-
"grad_norm": 3.1027884483337402,
|
| 127 |
-
"learning_rate": 4.923076923076924e-06,
|
| 128 |
-
"loss": 0.6097,
|
| 129 |
-
"step": 80
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.24615384615384617,
|
| 133 |
-
"eval_loss": 0.6625475883483887,
|
| 134 |
-
"eval_runtime": 1.1717,
|
| 135 |
-
"eval_samples_per_second": 97.294,
|
| 136 |
-
"eval_steps_per_second": 5.121,
|
| 137 |
-
"step": 80
|
| 138 |
-
},
|
| 139 |
-
{
|
| 140 |
-
"epoch": 0.27692307692307694,
|
| 141 |
-
"grad_norm": 3.233839511871338,
|
| 142 |
-
"learning_rate": 5.538461538461539e-06,
|
| 143 |
-
"loss": 0.6957,
|
| 144 |
-
"step": 90
|
| 145 |
-
},
|
| 146 |
-
{
|
| 147 |
-
"epoch": 0.27692307692307694,
|
| 148 |
-
"eval_loss": 0.6590337157249451,
|
| 149 |
-
"eval_runtime": 1.1664,
|
| 150 |
-
"eval_samples_per_second": 97.738,
|
| 151 |
-
"eval_steps_per_second": 5.144,
|
| 152 |
-
"step": 90
|
| 153 |
-
},
|
| 154 |
-
{
|
| 155 |
-
"epoch": 0.3076923076923077,
|
| 156 |
-
"grad_norm": 4.932614803314209,
|
| 157 |
-
"learning_rate": 6.153846153846155e-06,
|
| 158 |
-
"loss": 0.6994,
|
| 159 |
-
"step": 100
|
| 160 |
-
},
|
| 161 |
-
{
|
| 162 |
-
"epoch": 0.3076923076923077,
|
| 163 |
-
"eval_loss": 0.6592673659324646,
|
| 164 |
-
"eval_runtime": 1.1692,
|
| 165 |
-
"eval_samples_per_second": 97.502,
|
| 166 |
-
"eval_steps_per_second": 5.132,
|
| 167 |
-
"step": 100
|
| 168 |
-
}
|
| 169 |
-
],
|
| 170 |
-
"logging_steps": 10,
|
| 171 |
-
"max_steps": 3250,
|
| 172 |
-
"num_input_tokens_seen": 0,
|
| 173 |
-
"num_train_epochs": 10,
|
| 174 |
-
"save_steps": 50,
|
| 175 |
-
"total_flos": 6161848315412480.0,
|
| 176 |
-
"train_batch_size": 2,
|
| 177 |
-
"trial_name": null,
|
| 178 |
-
"trial_params": null
|
| 179 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
{checkpoint-100 → checkpoint-1000}/added_tokens.json
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/config.json
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/generation_config.json
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/model.safetensors.index.json
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/special_tokens_map.json
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/tokenizer.json
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/tokenizer.model
RENAMED
|
File without changes
|
{checkpoint-100 → checkpoint-1000}/tokenizer_config.json
RENAMED
|
File without changes
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,1529 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.076923076923077,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006153846153846154,
|
| 13 |
+
"eval_loss": 1.487821340560913,
|
| 14 |
+
"eval_runtime": 1.2374,
|
| 15 |
+
"eval_samples_per_second": 92.131,
|
| 16 |
+
"eval_steps_per_second": 4.849,
|
| 17 |
+
"step": 2
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03076923076923077,
|
| 21 |
+
"grad_norm": 20.007999420166016,
|
| 22 |
+
"learning_rate": 6.153846153846155e-07,
|
| 23 |
+
"loss": 1.4703,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03076923076923077,
|
| 28 |
+
"eval_loss": 1.4221965074539185,
|
| 29 |
+
"eval_runtime": 1.1701,
|
| 30 |
+
"eval_samples_per_second": 97.431,
|
| 31 |
+
"eval_steps_per_second": 5.128,
|
| 32 |
+
"step": 10
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06153846153846154,
|
| 36 |
+
"grad_norm": 29.657033920288086,
|
| 37 |
+
"learning_rate": 1.230769230769231e-06,
|
| 38 |
+
"loss": 1.4402,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06153846153846154,
|
| 43 |
+
"eval_loss": 1.238189935684204,
|
| 44 |
+
"eval_runtime": 1.1729,
|
| 45 |
+
"eval_samples_per_second": 97.195,
|
| 46 |
+
"eval_steps_per_second": 5.116,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.09230769230769231,
|
| 51 |
+
"grad_norm": 10.80688190460205,
|
| 52 |
+
"learning_rate": 1.8461538461538465e-06,
|
| 53 |
+
"loss": 1.0957,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09230769230769231,
|
| 58 |
+
"eval_loss": 0.908875048160553,
|
| 59 |
+
"eval_runtime": 1.1739,
|
| 60 |
+
"eval_samples_per_second": 97.111,
|
| 61 |
+
"eval_steps_per_second": 5.111,
|
| 62 |
+
"step": 30
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.12307692307692308,
|
| 66 |
+
"grad_norm": 12.914667129516602,
|
| 67 |
+
"learning_rate": 2.461538461538462e-06,
|
| 68 |
+
"loss": 0.8136,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.12307692307692308,
|
| 73 |
+
"eval_loss": 0.7436326146125793,
|
| 74 |
+
"eval_runtime": 1.1712,
|
| 75 |
+
"eval_samples_per_second": 97.34,
|
| 76 |
+
"eval_steps_per_second": 5.123,
|
| 77 |
+
"step": 40
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.15384615384615385,
|
| 81 |
+
"grad_norm": 5.206438064575195,
|
| 82 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 83 |
+
"loss": 0.7864,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.15384615384615385,
|
| 88 |
+
"eval_loss": 0.6924600601196289,
|
| 89 |
+
"eval_runtime": 1.1718,
|
| 90 |
+
"eval_samples_per_second": 97.287,
|
| 91 |
+
"eval_steps_per_second": 5.12,
|
| 92 |
+
"step": 50
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.18461538461538463,
|
| 96 |
+
"grad_norm": 5.670599460601807,
|
| 97 |
+
"learning_rate": 3.692307692307693e-06,
|
| 98 |
+
"loss": 0.622,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.18461538461538463,
|
| 103 |
+
"eval_loss": 0.6791077852249146,
|
| 104 |
+
"eval_runtime": 1.1678,
|
| 105 |
+
"eval_samples_per_second": 97.62,
|
| 106 |
+
"eval_steps_per_second": 5.138,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.2153846153846154,
|
| 111 |
+
"grad_norm": 3.6266868114471436,
|
| 112 |
+
"learning_rate": 4.307692307692308e-06,
|
| 113 |
+
"loss": 0.6672,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2153846153846154,
|
| 118 |
+
"eval_loss": 0.6631948947906494,
|
| 119 |
+
"eval_runtime": 1.1626,
|
| 120 |
+
"eval_samples_per_second": 98.06,
|
| 121 |
+
"eval_steps_per_second": 5.161,
|
| 122 |
+
"step": 70
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.24615384615384617,
|
| 126 |
+
"grad_norm": 3.1027884483337402,
|
| 127 |
+
"learning_rate": 4.923076923076924e-06,
|
| 128 |
+
"loss": 0.6097,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.24615384615384617,
|
| 133 |
+
"eval_loss": 0.6625475883483887,
|
| 134 |
+
"eval_runtime": 1.1717,
|
| 135 |
+
"eval_samples_per_second": 97.294,
|
| 136 |
+
"eval_steps_per_second": 5.121,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.27692307692307694,
|
| 141 |
+
"grad_norm": 3.233839511871338,
|
| 142 |
+
"learning_rate": 5.538461538461539e-06,
|
| 143 |
+
"loss": 0.6957,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.27692307692307694,
|
| 148 |
+
"eval_loss": 0.6590337157249451,
|
| 149 |
+
"eval_runtime": 1.1664,
|
| 150 |
+
"eval_samples_per_second": 97.738,
|
| 151 |
+
"eval_steps_per_second": 5.144,
|
| 152 |
+
"step": 90
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.3076923076923077,
|
| 156 |
+
"grad_norm": 4.932614803314209,
|
| 157 |
+
"learning_rate": 6.153846153846155e-06,
|
| 158 |
+
"loss": 0.6994,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.3076923076923077,
|
| 163 |
+
"eval_loss": 0.6592673659324646,
|
| 164 |
+
"eval_runtime": 1.1692,
|
| 165 |
+
"eval_samples_per_second": 97.502,
|
| 166 |
+
"eval_steps_per_second": 5.132,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.3384615384615385,
|
| 171 |
+
"grad_norm": 2.642303466796875,
|
| 172 |
+
"learning_rate": 6.76923076923077e-06,
|
| 173 |
+
"loss": 0.6501,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.3384615384615385,
|
| 178 |
+
"eval_loss": 0.6589992642402649,
|
| 179 |
+
"eval_runtime": 1.1669,
|
| 180 |
+
"eval_samples_per_second": 97.691,
|
| 181 |
+
"eval_steps_per_second": 5.142,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.36923076923076925,
|
| 186 |
+
"grad_norm": 3.45409893989563,
|
| 187 |
+
"learning_rate": 7.384615384615386e-06,
|
| 188 |
+
"loss": 0.7512,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.36923076923076925,
|
| 193 |
+
"eval_loss": 0.6568591594696045,
|
| 194 |
+
"eval_runtime": 1.1778,
|
| 195 |
+
"eval_samples_per_second": 96.79,
|
| 196 |
+
"eval_steps_per_second": 5.094,
|
| 197 |
+
"step": 120
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.4,
|
| 201 |
+
"grad_norm": 8.460351943969727,
|
| 202 |
+
"learning_rate": 8.000000000000001e-06,
|
| 203 |
+
"loss": 0.5996,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4,
|
| 208 |
+
"eval_loss": 0.6578297019004822,
|
| 209 |
+
"eval_runtime": 1.1683,
|
| 210 |
+
"eval_samples_per_second": 97.575,
|
| 211 |
+
"eval_steps_per_second": 5.136,
|
| 212 |
+
"step": 130
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4307692307692308,
|
| 216 |
+
"grad_norm": 2.4495465755462646,
|
| 217 |
+
"learning_rate": 8.615384615384617e-06,
|
| 218 |
+
"loss": 0.6498,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4307692307692308,
|
| 223 |
+
"eval_loss": 0.6588199138641357,
|
| 224 |
+
"eval_runtime": 1.1664,
|
| 225 |
+
"eval_samples_per_second": 97.739,
|
| 226 |
+
"eval_steps_per_second": 5.144,
|
| 227 |
+
"step": 140
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.46153846153846156,
|
| 231 |
+
"grad_norm": 2.6779162883758545,
|
| 232 |
+
"learning_rate": 9.230769230769232e-06,
|
| 233 |
+
"loss": 0.703,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.46153846153846156,
|
| 238 |
+
"eval_loss": 0.6626676321029663,
|
| 239 |
+
"eval_runtime": 1.1764,
|
| 240 |
+
"eval_samples_per_second": 96.904,
|
| 241 |
+
"eval_steps_per_second": 5.1,
|
| 242 |
+
"step": 150
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.49230769230769234,
|
| 246 |
+
"grad_norm": 8.147880554199219,
|
| 247 |
+
"learning_rate": 9.846153846153848e-06,
|
| 248 |
+
"loss": 0.6649,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.49230769230769234,
|
| 253 |
+
"eval_loss": 0.6644802689552307,
|
| 254 |
+
"eval_runtime": 1.1701,
|
| 255 |
+
"eval_samples_per_second": 97.424,
|
| 256 |
+
"eval_steps_per_second": 5.128,
|
| 257 |
+
"step": 160
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.5230769230769231,
|
| 261 |
+
"grad_norm": 2.7410728931427,
|
| 262 |
+
"learning_rate": 1.0461538461538463e-05,
|
| 263 |
+
"loss": 0.6704,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.5230769230769231,
|
| 268 |
+
"eval_loss": 0.6591455340385437,
|
| 269 |
+
"eval_runtime": 1.1722,
|
| 270 |
+
"eval_samples_per_second": 97.253,
|
| 271 |
+
"eval_steps_per_second": 5.119,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.5538461538461539,
|
| 276 |
+
"grad_norm": 6.886598587036133,
|
| 277 |
+
"learning_rate": 1.1076923076923079e-05,
|
| 278 |
+
"loss": 0.6073,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.5538461538461539,
|
| 283 |
+
"eval_loss": 0.6617754697799683,
|
| 284 |
+
"eval_runtime": 1.1682,
|
| 285 |
+
"eval_samples_per_second": 97.588,
|
| 286 |
+
"eval_steps_per_second": 5.136,
|
| 287 |
+
"step": 180
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.5846153846153846,
|
| 291 |
+
"grad_norm": 4.26249361038208,
|
| 292 |
+
"learning_rate": 1.1692307692307694e-05,
|
| 293 |
+
"loss": 0.7556,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.5846153846153846,
|
| 298 |
+
"eval_loss": 0.6659030318260193,
|
| 299 |
+
"eval_runtime": 1.1721,
|
| 300 |
+
"eval_samples_per_second": 97.265,
|
| 301 |
+
"eval_steps_per_second": 5.119,
|
| 302 |
+
"step": 190
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.6153846153846154,
|
| 306 |
+
"grad_norm": 4.204222202301025,
|
| 307 |
+
"learning_rate": 1.230769230769231e-05,
|
| 308 |
+
"loss": 0.6583,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.6153846153846154,
|
| 313 |
+
"eval_loss": 0.6660778522491455,
|
| 314 |
+
"eval_runtime": 1.1728,
|
| 315 |
+
"eval_samples_per_second": 97.201,
|
| 316 |
+
"eval_steps_per_second": 5.116,
|
| 317 |
+
"step": 200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6461538461538462,
|
| 321 |
+
"grad_norm": 3.154658079147339,
|
| 322 |
+
"learning_rate": 1.2923076923076925e-05,
|
| 323 |
+
"loss": 0.7288,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6461538461538462,
|
| 328 |
+
"eval_loss": 0.6642422676086426,
|
| 329 |
+
"eval_runtime": 1.166,
|
| 330 |
+
"eval_samples_per_second": 97.772,
|
| 331 |
+
"eval_steps_per_second": 5.146,
|
| 332 |
+
"step": 210
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.676923076923077,
|
| 336 |
+
"grad_norm": 3.248359203338623,
|
| 337 |
+
"learning_rate": 1.353846153846154e-05,
|
| 338 |
+
"loss": 0.6702,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.676923076923077,
|
| 343 |
+
"eval_loss": 0.6680281758308411,
|
| 344 |
+
"eval_runtime": 1.1691,
|
| 345 |
+
"eval_samples_per_second": 97.507,
|
| 346 |
+
"eval_steps_per_second": 5.132,
|
| 347 |
+
"step": 220
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.7076923076923077,
|
| 351 |
+
"grad_norm": 2.5507216453552246,
|
| 352 |
+
"learning_rate": 1.4153846153846156e-05,
|
| 353 |
+
"loss": 0.66,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.7076923076923077,
|
| 358 |
+
"eval_loss": 0.6713310480117798,
|
| 359 |
+
"eval_runtime": 1.167,
|
| 360 |
+
"eval_samples_per_second": 97.683,
|
| 361 |
+
"eval_steps_per_second": 5.141,
|
| 362 |
+
"step": 230
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.7384615384615385,
|
| 366 |
+
"grad_norm": 2.1180007457733154,
|
| 367 |
+
"learning_rate": 1.4769230769230772e-05,
|
| 368 |
+
"loss": 0.6172,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.7384615384615385,
|
| 373 |
+
"eval_loss": 0.6759226322174072,
|
| 374 |
+
"eval_runtime": 1.1758,
|
| 375 |
+
"eval_samples_per_second": 96.954,
|
| 376 |
+
"eval_steps_per_second": 5.103,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.7692307692307693,
|
| 381 |
+
"grad_norm": 2.8318371772766113,
|
| 382 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 383 |
+
"loss": 0.5954,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.7692307692307693,
|
| 388 |
+
"eval_loss": 0.6854314208030701,
|
| 389 |
+
"eval_runtime": 1.1698,
|
| 390 |
+
"eval_samples_per_second": 97.457,
|
| 391 |
+
"eval_steps_per_second": 5.129,
|
| 392 |
+
"step": 250
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.8,
|
| 396 |
+
"grad_norm": 3.519829034805298,
|
| 397 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 398 |
+
"loss": 0.6639,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.8,
|
| 403 |
+
"eval_loss": 0.6849851012229919,
|
| 404 |
+
"eval_runtime": 1.1732,
|
| 405 |
+
"eval_samples_per_second": 97.173,
|
| 406 |
+
"eval_steps_per_second": 5.114,
|
| 407 |
+
"step": 260
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.8307692307692308,
|
| 411 |
+
"grad_norm": 3.7909388542175293,
|
| 412 |
+
"learning_rate": 1.6615384615384618e-05,
|
| 413 |
+
"loss": 0.6892,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.8307692307692308,
|
| 418 |
+
"eval_loss": 0.6863855123519897,
|
| 419 |
+
"eval_runtime": 1.1689,
|
| 420 |
+
"eval_samples_per_second": 97.53,
|
| 421 |
+
"eval_steps_per_second": 5.133,
|
| 422 |
+
"step": 270
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8615384615384616,
|
| 426 |
+
"grad_norm": 8.278326988220215,
|
| 427 |
+
"learning_rate": 1.7230769230769234e-05,
|
| 428 |
+
"loss": 0.6014,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8615384615384616,
|
| 433 |
+
"eval_loss": 0.6890373826026917,
|
| 434 |
+
"eval_runtime": 1.1722,
|
| 435 |
+
"eval_samples_per_second": 97.251,
|
| 436 |
+
"eval_steps_per_second": 5.118,
|
| 437 |
+
"step": 280
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8923076923076924,
|
| 441 |
+
"grad_norm": 2.616001605987549,
|
| 442 |
+
"learning_rate": 1.784615384615385e-05,
|
| 443 |
+
"loss": 0.6775,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.8923076923076924,
|
| 448 |
+
"eval_loss": 0.6797110438346863,
|
| 449 |
+
"eval_runtime": 1.1888,
|
| 450 |
+
"eval_samples_per_second": 95.897,
|
| 451 |
+
"eval_steps_per_second": 5.047,
|
| 452 |
+
"step": 290
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.9230769230769231,
|
| 456 |
+
"grad_norm": 2.4450528621673584,
|
| 457 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 458 |
+
"loss": 0.7434,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.9230769230769231,
|
| 463 |
+
"eval_loss": 0.6779448986053467,
|
| 464 |
+
"eval_runtime": 1.1722,
|
| 465 |
+
"eval_samples_per_second": 97.254,
|
| 466 |
+
"eval_steps_per_second": 5.119,
|
| 467 |
+
"step": 300
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.9538461538461539,
|
| 471 |
+
"grad_norm": 4.121164798736572,
|
| 472 |
+
"learning_rate": 1.907692307692308e-05,
|
| 473 |
+
"loss": 0.6413,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.9538461538461539,
|
| 478 |
+
"eval_loss": 0.6819297671318054,
|
| 479 |
+
"eval_runtime": 1.1683,
|
| 480 |
+
"eval_samples_per_second": 97.574,
|
| 481 |
+
"eval_steps_per_second": 5.135,
|
| 482 |
+
"step": 310
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.9846153846153847,
|
| 486 |
+
"grad_norm": 2.057229995727539,
|
| 487 |
+
"learning_rate": 1.9692307692307696e-05,
|
| 488 |
+
"loss": 0.6511,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.9846153846153847,
|
| 493 |
+
"eval_loss": 0.6861481666564941,
|
| 494 |
+
"eval_runtime": 1.1699,
|
| 495 |
+
"eval_samples_per_second": 97.444,
|
| 496 |
+
"eval_steps_per_second": 5.129,
|
| 497 |
+
"step": 320
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.0153846153846153,
|
| 501 |
+
"grad_norm": 1.7295210361480713,
|
| 502 |
+
"learning_rate": 1.9999855802751384e-05,
|
| 503 |
+
"loss": 0.5872,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.0153846153846153,
|
| 508 |
+
"eval_loss": 0.7481781244277954,
|
| 509 |
+
"eval_runtime": 1.172,
|
| 510 |
+
"eval_samples_per_second": 97.273,
|
| 511 |
+
"eval_steps_per_second": 5.12,
|
| 512 |
+
"step": 330
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.0461538461538462,
|
| 516 |
+
"grad_norm": 3.6910336017608643,
|
| 517 |
+
"learning_rate": 1.9998702249713747e-05,
|
| 518 |
+
"loss": 0.438,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.0461538461538462,
|
| 523 |
+
"eval_loss": 0.7471169829368591,
|
| 524 |
+
"eval_runtime": 1.175,
|
| 525 |
+
"eval_samples_per_second": 97.019,
|
| 526 |
+
"eval_steps_per_second": 5.106,
|
| 527 |
+
"step": 340
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.0769230769230769,
|
| 531 |
+
"grad_norm": 3.9220831394195557,
|
| 532 |
+
"learning_rate": 1.9996395276708856e-05,
|
| 533 |
+
"loss": 0.4141,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0769230769230769,
|
| 538 |
+
"eval_loss": 0.7425897121429443,
|
| 539 |
+
"eval_runtime": 1.1888,
|
| 540 |
+
"eval_samples_per_second": 95.893,
|
| 541 |
+
"eval_steps_per_second": 5.047,
|
| 542 |
+
"step": 350
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.1076923076923078,
|
| 546 |
+
"grad_norm": 2.1136879920959473,
|
| 547 |
+
"learning_rate": 1.9992935149862116e-05,
|
| 548 |
+
"loss": 0.4173,
|
| 549 |
+
"step": 360
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1076923076923078,
|
| 553 |
+
"eval_loss": 0.7477542161941528,
|
| 554 |
+
"eval_runtime": 1.1705,
|
| 555 |
+
"eval_samples_per_second": 97.398,
|
| 556 |
+
"eval_steps_per_second": 5.126,
|
| 557 |
+
"step": 360
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.1384615384615384,
|
| 561 |
+
"grad_norm": 2.187299966812134,
|
| 562 |
+
"learning_rate": 1.998832226832327e-05,
|
| 563 |
+
"loss": 0.4242,
|
| 564 |
+
"step": 370
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.1384615384615384,
|
| 568 |
+
"eval_loss": 0.7475994229316711,
|
| 569 |
+
"eval_runtime": 1.1683,
|
| 570 |
+
"eval_samples_per_second": 97.582,
|
| 571 |
+
"eval_steps_per_second": 5.136,
|
| 572 |
+
"step": 370
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.1692307692307693,
|
| 576 |
+
"grad_norm": 2.683750629425049,
|
| 577 |
+
"learning_rate": 1.9982557164220335e-05,
|
| 578 |
+
"loss": 0.4413,
|
| 579 |
+
"step": 380
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.1692307692307693,
|
| 583 |
+
"eval_loss": 0.7552060484886169,
|
| 584 |
+
"eval_runtime": 1.1689,
|
| 585 |
+
"eval_samples_per_second": 97.529,
|
| 586 |
+
"eval_steps_per_second": 5.133,
|
| 587 |
+
"step": 380
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.2,
|
| 591 |
+
"grad_norm": 3.312657117843628,
|
| 592 |
+
"learning_rate": 1.9975640502598243e-05,
|
| 593 |
+
"loss": 0.3361,
|
| 594 |
+
"step": 390
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.2,
|
| 598 |
+
"eval_loss": 0.7630029916763306,
|
| 599 |
+
"eval_runtime": 1.1689,
|
| 600 |
+
"eval_samples_per_second": 97.529,
|
| 601 |
+
"eval_steps_per_second": 5.133,
|
| 602 |
+
"step": 390
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.2307692307692308,
|
| 606 |
+
"grad_norm": 2.3290536403656006,
|
| 607 |
+
"learning_rate": 1.9967573081342103e-05,
|
| 608 |
+
"loss": 0.4174,
|
| 609 |
+
"step": 400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.2307692307692308,
|
| 613 |
+
"eval_loss": 0.7669293880462646,
|
| 614 |
+
"eval_runtime": 1.1711,
|
| 615 |
+
"eval_samples_per_second": 97.345,
|
| 616 |
+
"eval_steps_per_second": 5.123,
|
| 617 |
+
"step": 400
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.2615384615384615,
|
| 621 |
+
"grad_norm": 2.4483697414398193,
|
| 622 |
+
"learning_rate": 1.9958355831085155e-05,
|
| 623 |
+
"loss": 0.4045,
|
| 624 |
+
"step": 410
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.2615384615384615,
|
| 628 |
+
"eval_loss": 0.7652024626731873,
|
| 629 |
+
"eval_runtime": 1.1697,
|
| 630 |
+
"eval_samples_per_second": 97.461,
|
| 631 |
+
"eval_steps_per_second": 5.13,
|
| 632 |
+
"step": 410
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.2923076923076924,
|
| 636 |
+
"grad_norm": 1.703616976737976,
|
| 637 |
+
"learning_rate": 1.9947989815101444e-05,
|
| 638 |
+
"loss": 0.4292,
|
| 639 |
+
"step": 420
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.2923076923076924,
|
| 643 |
+
"eval_loss": 0.7553891539573669,
|
| 644 |
+
"eval_runtime": 1.1719,
|
| 645 |
+
"eval_samples_per_second": 97.277,
|
| 646 |
+
"eval_steps_per_second": 5.12,
|
| 647 |
+
"step": 420
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.323076923076923,
|
| 651 |
+
"grad_norm": 1.6245399713516235,
|
| 652 |
+
"learning_rate": 1.9936476229183133e-05,
|
| 653 |
+
"loss": 0.4164,
|
| 654 |
+
"step": 430
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.323076923076923,
|
| 658 |
+
"eval_loss": 0.7711736559867859,
|
| 659 |
+
"eval_runtime": 1.1776,
|
| 660 |
+
"eval_samples_per_second": 96.806,
|
| 661 |
+
"eval_steps_per_second": 5.095,
|
| 662 |
+
"step": 430
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.353846153846154,
|
| 666 |
+
"grad_norm": 1.9511611461639404,
|
| 667 |
+
"learning_rate": 1.992381640150257e-05,
|
| 668 |
+
"loss": 0.4337,
|
| 669 |
+
"step": 440
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.353846153846154,
|
| 673 |
+
"eval_loss": 0.7438372373580933,
|
| 674 |
+
"eval_runtime": 1.1692,
|
| 675 |
+
"eval_samples_per_second": 97.5,
|
| 676 |
+
"eval_steps_per_second": 5.132,
|
| 677 |
+
"step": 440
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.3846153846153846,
|
| 681 |
+
"grad_norm": 2.466737747192383,
|
| 682 |
+
"learning_rate": 1.9910011792459086e-05,
|
| 683 |
+
"loss": 0.3712,
|
| 684 |
+
"step": 450
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.3846153846153846,
|
| 688 |
+
"eval_loss": 0.8045614361763,
|
| 689 |
+
"eval_runtime": 1.1689,
|
| 690 |
+
"eval_samples_per_second": 97.53,
|
| 691 |
+
"eval_steps_per_second": 5.133,
|
| 692 |
+
"step": 450
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.4153846153846155,
|
| 696 |
+
"grad_norm": 2.473851203918457,
|
| 697 |
+
"learning_rate": 1.9895063994510512e-05,
|
| 698 |
+
"loss": 0.516,
|
| 699 |
+
"step": 460
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.4153846153846155,
|
| 703 |
+
"eval_loss": 0.7463352084159851,
|
| 704 |
+
"eval_runtime": 1.1697,
|
| 705 |
+
"eval_samples_per_second": 97.464,
|
| 706 |
+
"eval_steps_per_second": 5.13,
|
| 707 |
+
"step": 460
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.4461538461538461,
|
| 711 |
+
"grad_norm": 2.757955312728882,
|
| 712 |
+
"learning_rate": 1.9878974731989487e-05,
|
| 713 |
+
"loss": 0.4965,
|
| 714 |
+
"step": 470
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.4461538461538461,
|
| 718 |
+
"eval_loss": 0.7571713924407959,
|
| 719 |
+
"eval_runtime": 1.1688,
|
| 720 |
+
"eval_samples_per_second": 97.534,
|
| 721 |
+
"eval_steps_per_second": 5.133,
|
| 722 |
+
"step": 470
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.476923076923077,
|
| 726 |
+
"grad_norm": 3.9884915351867676,
|
| 727 |
+
"learning_rate": 1.9861745860904538e-05,
|
| 728 |
+
"loss": 0.3973,
|
| 729 |
+
"step": 480
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.476923076923077,
|
| 733 |
+
"eval_loss": 0.7763081789016724,
|
| 734 |
+
"eval_runtime": 1.168,
|
| 735 |
+
"eval_samples_per_second": 97.605,
|
| 736 |
+
"eval_steps_per_second": 5.137,
|
| 737 |
+
"step": 480
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5076923076923077,
|
| 741 |
+
"grad_norm": 1.9474631547927856,
|
| 742 |
+
"learning_rate": 1.9843379368725978e-05,
|
| 743 |
+
"loss": 0.4625,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5076923076923077,
|
| 748 |
+
"eval_loss": 0.7595367431640625,
|
| 749 |
+
"eval_runtime": 1.1708,
|
| 750 |
+
"eval_samples_per_second": 97.37,
|
| 751 |
+
"eval_steps_per_second": 5.125,
|
| 752 |
+
"step": 490
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5384615384615383,
|
| 756 |
+
"grad_norm": 2.2907869815826416,
|
| 757 |
+
"learning_rate": 1.9823877374156647e-05,
|
| 758 |
+
"loss": 0.4371,
|
| 759 |
+
"step": 500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5384615384615383,
|
| 763 |
+
"eval_loss": 0.762865424156189,
|
| 764 |
+
"eval_runtime": 1.1766,
|
| 765 |
+
"eval_samples_per_second": 96.891,
|
| 766 |
+
"eval_steps_per_second": 5.1,
|
| 767 |
+
"step": 500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.5692307692307692,
|
| 771 |
+
"grad_norm": 3.067608594894409,
|
| 772 |
+
"learning_rate": 1.9803242126887496e-05,
|
| 773 |
+
"loss": 0.45,
|
| 774 |
+
"step": 510
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.5692307692307692,
|
| 778 |
+
"eval_loss": 0.7638718485832214,
|
| 779 |
+
"eval_runtime": 1.1684,
|
| 780 |
+
"eval_samples_per_second": 97.568,
|
| 781 |
+
"eval_steps_per_second": 5.135,
|
| 782 |
+
"step": 510
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.6,
|
| 786 |
+
"grad_norm": 2.1113197803497314,
|
| 787 |
+
"learning_rate": 1.9781476007338058e-05,
|
| 788 |
+
"loss": 0.4524,
|
| 789 |
+
"step": 520
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.6,
|
| 793 |
+
"eval_loss": 0.7560412883758545,
|
| 794 |
+
"eval_runtime": 1.1719,
|
| 795 |
+
"eval_samples_per_second": 97.278,
|
| 796 |
+
"eval_steps_per_second": 5.12,
|
| 797 |
+
"step": 520
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.6307692307692307,
|
| 801 |
+
"grad_norm": 1.889585256576538,
|
| 802 |
+
"learning_rate": 1.9758581526381878e-05,
|
| 803 |
+
"loss": 0.4287,
|
| 804 |
+
"step": 530
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.6307692307692307,
|
| 808 |
+
"eval_loss": 0.7591350674629211,
|
| 809 |
+
"eval_runtime": 1.1693,
|
| 810 |
+
"eval_samples_per_second": 97.495,
|
| 811 |
+
"eval_steps_per_second": 5.131,
|
| 812 |
+
"step": 530
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.6615384615384614,
|
| 816 |
+
"grad_norm": 2.519418716430664,
|
| 817 |
+
"learning_rate": 1.973456132505684e-05,
|
| 818 |
+
"loss": 0.4603,
|
| 819 |
+
"step": 540
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.6615384615384614,
|
| 823 |
+
"eval_loss": 0.7536805868148804,
|
| 824 |
+
"eval_runtime": 1.1799,
|
| 825 |
+
"eval_samples_per_second": 96.617,
|
| 826 |
+
"eval_steps_per_second": 5.085,
|
| 827 |
+
"step": 540
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.6923076923076923,
|
| 831 |
+
"grad_norm": 2.7455596923828125,
|
| 832 |
+
"learning_rate": 1.9709418174260523e-05,
|
| 833 |
+
"loss": 0.4293,
|
| 834 |
+
"step": 550
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.6923076923076923,
|
| 838 |
+
"eval_loss": 0.763169527053833,
|
| 839 |
+
"eval_runtime": 1.1696,
|
| 840 |
+
"eval_samples_per_second": 97.47,
|
| 841 |
+
"eval_steps_per_second": 5.13,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7230769230769232,
|
| 846 |
+
"grad_norm": 1.9823411703109741,
|
| 847 |
+
"learning_rate": 1.9683154974430544e-05,
|
| 848 |
+
"loss": 0.4643,
|
| 849 |
+
"step": 560
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7230769230769232,
|
| 853 |
+
"eval_loss": 0.7607314586639404,
|
| 854 |
+
"eval_runtime": 1.1712,
|
| 855 |
+
"eval_samples_per_second": 97.336,
|
| 856 |
+
"eval_steps_per_second": 5.123,
|
| 857 |
+
"step": 560
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.7538461538461538,
|
| 861 |
+
"grad_norm": 2.2437684535980225,
|
| 862 |
+
"learning_rate": 1.965577475520999e-05,
|
| 863 |
+
"loss": 0.452,
|
| 864 |
+
"step": 570
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.7538461538461538,
|
| 868 |
+
"eval_loss": 0.7615717649459839,
|
| 869 |
+
"eval_runtime": 1.1946,
|
| 870 |
+
"eval_samples_per_second": 95.425,
|
| 871 |
+
"eval_steps_per_second": 5.022,
|
| 872 |
+
"step": 570
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.7846153846153845,
|
| 876 |
+
"grad_norm": 3.0740182399749756,
|
| 877 |
+
"learning_rate": 1.962728067509791e-05,
|
| 878 |
+
"loss": 0.4885,
|
| 879 |
+
"step": 580
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.7846153846153845,
|
| 883 |
+
"eval_loss": 0.7764689326286316,
|
| 884 |
+
"eval_runtime": 1.1714,
|
| 885 |
+
"eval_samples_per_second": 97.32,
|
| 886 |
+
"eval_steps_per_second": 5.122,
|
| 887 |
+
"step": 580
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.8153846153846154,
|
| 891 |
+
"grad_norm": 2.050313711166382,
|
| 892 |
+
"learning_rate": 1.9597676021084962e-05,
|
| 893 |
+
"loss": 0.4992,
|
| 894 |
+
"step": 590
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.8153846153846154,
|
| 898 |
+
"eval_loss": 0.7457339763641357,
|
| 899 |
+
"eval_runtime": 1.1747,
|
| 900 |
+
"eval_samples_per_second": 97.045,
|
| 901 |
+
"eval_steps_per_second": 5.108,
|
| 902 |
+
"step": 590
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.8461538461538463,
|
| 906 |
+
"grad_norm": 2.349470853805542,
|
| 907 |
+
"learning_rate": 1.9566964208274254e-05,
|
| 908 |
+
"loss": 0.3963,
|
| 909 |
+
"step": 600
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.8461538461538463,
|
| 913 |
+
"eval_loss": 0.7712545990943909,
|
| 914 |
+
"eval_runtime": 1.17,
|
| 915 |
+
"eval_samples_per_second": 97.433,
|
| 916 |
+
"eval_steps_per_second": 5.128,
|
| 917 |
+
"step": 600
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.876923076923077,
|
| 921 |
+
"grad_norm": 1.947912573814392,
|
| 922 |
+
"learning_rate": 1.9535148779487365e-05,
|
| 923 |
+
"loss": 0.4388,
|
| 924 |
+
"step": 610
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.876923076923077,
|
| 928 |
+
"eval_loss": 0.772519052028656,
|
| 929 |
+
"eval_runtime": 1.1717,
|
| 930 |
+
"eval_samples_per_second": 97.296,
|
| 931 |
+
"eval_steps_per_second": 5.121,
|
| 932 |
+
"step": 610
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.9076923076923076,
|
| 936 |
+
"grad_norm": 1.8837225437164307,
|
| 937 |
+
"learning_rate": 1.9502233404855672e-05,
|
| 938 |
+
"loss": 0.3965,
|
| 939 |
+
"step": 620
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.9076923076923076,
|
| 943 |
+
"eval_loss": 0.7594517469406128,
|
| 944 |
+
"eval_runtime": 1.2215,
|
| 945 |
+
"eval_samples_per_second": 93.326,
|
| 946 |
+
"eval_steps_per_second": 4.912,
|
| 947 |
+
"step": 620
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9384615384615385,
|
| 951 |
+
"grad_norm": 1.8398792743682861,
|
| 952 |
+
"learning_rate": 1.946822188139696e-05,
|
| 953 |
+
"loss": 0.3898,
|
| 954 |
+
"step": 630
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.9384615384615385,
|
| 958 |
+
"eval_loss": 0.7748451828956604,
|
| 959 |
+
"eval_runtime": 1.2222,
|
| 960 |
+
"eval_samples_per_second": 93.273,
|
| 961 |
+
"eval_steps_per_second": 4.909,
|
| 962 |
+
"step": 630
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.9692307692307693,
|
| 966 |
+
"grad_norm": 2.3971219062805176,
|
| 967 |
+
"learning_rate": 1.9433118132577432e-05,
|
| 968 |
+
"loss": 0.4253,
|
| 969 |
+
"step": 640
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.9692307692307693,
|
| 973 |
+
"eval_loss": 0.7729252576828003,
|
| 974 |
+
"eval_runtime": 1.1743,
|
| 975 |
+
"eval_samples_per_second": 97.081,
|
| 976 |
+
"eval_steps_per_second": 5.11,
|
| 977 |
+
"step": 640
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.0,
|
| 981 |
+
"grad_norm": 1.657328724861145,
|
| 982 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 983 |
+
"loss": 0.4898,
|
| 984 |
+
"step": 650
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.0,
|
| 988 |
+
"eval_loss": 0.7602188587188721,
|
| 989 |
+
"eval_runtime": 1.1738,
|
| 990 |
+
"eval_samples_per_second": 97.117,
|
| 991 |
+
"eval_steps_per_second": 5.111,
|
| 992 |
+
"step": 650
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.0307692307692307,
|
| 996 |
+
"grad_norm": 1.3759773969650269,
|
| 997 |
+
"learning_rate": 1.935965028223259e-05,
|
| 998 |
+
"loss": 0.2182,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.0307692307692307,
|
| 1003 |
+
"eval_loss": 0.8499692678451538,
|
| 1004 |
+
"eval_runtime": 1.1719,
|
| 1005 |
+
"eval_samples_per_second": 97.276,
|
| 1006 |
+
"eval_steps_per_second": 5.12,
|
| 1007 |
+
"step": 660
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.0615384615384613,
|
| 1011 |
+
"grad_norm": 1.4757074117660522,
|
| 1012 |
+
"learning_rate": 1.932129465573568e-05,
|
| 1013 |
+
"loss": 0.2196,
|
| 1014 |
+
"step": 670
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.0615384615384613,
|
| 1018 |
+
"eval_loss": 0.866599977016449,
|
| 1019 |
+
"eval_runtime": 1.1739,
|
| 1020 |
+
"eval_samples_per_second": 97.115,
|
| 1021 |
+
"eval_steps_per_second": 5.111,
|
| 1022 |
+
"step": 670
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.0923076923076924,
|
| 1026 |
+
"grad_norm": 1.2500200271606445,
|
| 1027 |
+
"learning_rate": 1.9281863752957095e-05,
|
| 1028 |
+
"loss": 0.1772,
|
| 1029 |
+
"step": 680
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.0923076923076924,
|
| 1033 |
+
"eval_loss": 0.8642630577087402,
|
| 1034 |
+
"eval_runtime": 1.1753,
|
| 1035 |
+
"eval_samples_per_second": 96.999,
|
| 1036 |
+
"eval_steps_per_second": 5.105,
|
| 1037 |
+
"step": 680
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.123076923076923,
|
| 1041 |
+
"grad_norm": 2.561828851699829,
|
| 1042 |
+
"learning_rate": 1.92413621225262e-05,
|
| 1043 |
+
"loss": 0.2014,
|
| 1044 |
+
"step": 690
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.123076923076923,
|
| 1048 |
+
"eval_loss": 0.8872668743133545,
|
| 1049 |
+
"eval_runtime": 1.1761,
|
| 1050 |
+
"eval_samples_per_second": 96.933,
|
| 1051 |
+
"eval_steps_per_second": 5.102,
|
| 1052 |
+
"step": 690
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.1538461538461537,
|
| 1056 |
+
"grad_norm": 1.8431521654129028,
|
| 1057 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 1058 |
+
"loss": 0.2245,
|
| 1059 |
+
"step": 700
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.1538461538461537,
|
| 1063 |
+
"eval_loss": 0.8799911737442017,
|
| 1064 |
+
"eval_runtime": 1.1766,
|
| 1065 |
+
"eval_samples_per_second": 96.89,
|
| 1066 |
+
"eval_steps_per_second": 5.099,
|
| 1067 |
+
"step": 700
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.184615384615385,
|
| 1071 |
+
"grad_norm": 1.4718531370162964,
|
| 1072 |
+
"learning_rate": 1.915716549026541e-05,
|
| 1073 |
+
"loss": 0.1911,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.184615384615385,
|
| 1078 |
+
"eval_loss": 0.8806982636451721,
|
| 1079 |
+
"eval_runtime": 1.1696,
|
| 1080 |
+
"eval_samples_per_second": 97.473,
|
| 1081 |
+
"eval_steps_per_second": 5.13,
|
| 1082 |
+
"step": 710
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.2153846153846155,
|
| 1086 |
+
"grad_norm": 2.061520576477051,
|
| 1087 |
+
"learning_rate": 1.9113480201103658e-05,
|
| 1088 |
+
"loss": 0.232,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.2153846153846155,
|
| 1093 |
+
"eval_loss": 0.8812103271484375,
|
| 1094 |
+
"eval_runtime": 1.1676,
|
| 1095 |
+
"eval_samples_per_second": 97.636,
|
| 1096 |
+
"eval_steps_per_second": 5.139,
|
| 1097 |
+
"step": 720
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.246153846153846,
|
| 1101 |
+
"grad_norm": 2.530285596847534,
|
| 1102 |
+
"learning_rate": 1.9068743608505454e-05,
|
| 1103 |
+
"loss": 0.2232,
|
| 1104 |
+
"step": 730
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.246153846153846,
|
| 1108 |
+
"eval_loss": 0.8682798147201538,
|
| 1109 |
+
"eval_runtime": 1.1687,
|
| 1110 |
+
"eval_samples_per_second": 97.545,
|
| 1111 |
+
"eval_steps_per_second": 5.134,
|
| 1112 |
+
"step": 730
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.276923076923077,
|
| 1116 |
+
"grad_norm": 1.8433207273483276,
|
| 1117 |
+
"learning_rate": 1.902296087314845e-05,
|
| 1118 |
+
"loss": 0.2315,
|
| 1119 |
+
"step": 740
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.276923076923077,
|
| 1123 |
+
"eval_loss": 0.8747490048408508,
|
| 1124 |
+
"eval_runtime": 1.1702,
|
| 1125 |
+
"eval_samples_per_second": 97.421,
|
| 1126 |
+
"eval_steps_per_second": 5.127,
|
| 1127 |
+
"step": 740
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.3076923076923075,
|
| 1131 |
+
"grad_norm": 2.1529664993286133,
|
| 1132 |
+
"learning_rate": 1.8976137276390145e-05,
|
| 1133 |
+
"loss": 0.239,
|
| 1134 |
+
"step": 750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.3076923076923075,
|
| 1138 |
+
"eval_loss": 0.8778530359268188,
|
| 1139 |
+
"eval_runtime": 1.1674,
|
| 1140 |
+
"eval_samples_per_second": 97.656,
|
| 1141 |
+
"eval_steps_per_second": 5.14,
|
| 1142 |
+
"step": 750
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.3384615384615386,
|
| 1146 |
+
"grad_norm": 1.839979887008667,
|
| 1147 |
+
"learning_rate": 1.892827821965864e-05,
|
| 1148 |
+
"loss": 0.2228,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.3384615384615386,
|
| 1153 |
+
"eval_loss": 0.8769873976707458,
|
| 1154 |
+
"eval_runtime": 1.1718,
|
| 1155 |
+
"eval_samples_per_second": 97.288,
|
| 1156 |
+
"eval_steps_per_second": 5.12,
|
| 1157 |
+
"step": 760
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.3692307692307693,
|
| 1161 |
+
"grad_norm": 1.3814674615859985,
|
| 1162 |
+
"learning_rate": 1.8879389223829592e-05,
|
| 1163 |
+
"loss": 0.1999,
|
| 1164 |
+
"step": 770
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.3692307692307693,
|
| 1168 |
+
"eval_loss": 0.8731983304023743,
|
| 1169 |
+
"eval_runtime": 1.1723,
|
| 1170 |
+
"eval_samples_per_second": 97.242,
|
| 1171 |
+
"eval_steps_per_second": 5.118,
|
| 1172 |
+
"step": 770
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.4,
|
| 1176 |
+
"grad_norm": 1.5382564067840576,
|
| 1177 |
+
"learning_rate": 1.8829475928589272e-05,
|
| 1178 |
+
"loss": 0.1888,
|
| 1179 |
+
"step": 780
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.4,
|
| 1183 |
+
"eval_loss": 0.8794592022895813,
|
| 1184 |
+
"eval_runtime": 1.1678,
|
| 1185 |
+
"eval_samples_per_second": 97.622,
|
| 1186 |
+
"eval_steps_per_second": 5.138,
|
| 1187 |
+
"step": 780
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.430769230769231,
|
| 1191 |
+
"grad_norm": 2.371277093887329,
|
| 1192 |
+
"learning_rate": 1.8778544091784047e-05,
|
| 1193 |
+
"loss": 0.2293,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.430769230769231,
|
| 1198 |
+
"eval_loss": 0.8893064260482788,
|
| 1199 |
+
"eval_runtime": 1.1684,
|
| 1200 |
+
"eval_samples_per_second": 97.567,
|
| 1201 |
+
"eval_steps_per_second": 5.135,
|
| 1202 |
+
"step": 790
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.4615384615384617,
|
| 1206 |
+
"grad_norm": 1.4892629384994507,
|
| 1207 |
+
"learning_rate": 1.8726599588756144e-05,
|
| 1208 |
+
"loss": 0.2326,
|
| 1209 |
+
"step": 800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.4615384615384617,
|
| 1213 |
+
"eval_loss": 0.8676194548606873,
|
| 1214 |
+
"eval_runtime": 1.1699,
|
| 1215 |
+
"eval_samples_per_second": 97.441,
|
| 1216 |
+
"eval_steps_per_second": 5.128,
|
| 1217 |
+
"step": 800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.4923076923076923,
|
| 1221 |
+
"grad_norm": 1.6047152280807495,
|
| 1222 |
+
"learning_rate": 1.8673648411665895e-05,
|
| 1223 |
+
"loss": 0.1986,
|
| 1224 |
+
"step": 810
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.4923076923076923,
|
| 1228 |
+
"eval_loss": 0.8942193984985352,
|
| 1229 |
+
"eval_runtime": 1.17,
|
| 1230 |
+
"eval_samples_per_second": 97.433,
|
| 1231 |
+
"eval_steps_per_second": 5.128,
|
| 1232 |
+
"step": 810
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.523076923076923,
|
| 1236 |
+
"grad_norm": 2.106461524963379,
|
| 1237 |
+
"learning_rate": 1.8619696668800494e-05,
|
| 1238 |
+
"loss": 0.238,
|
| 1239 |
+
"step": 820
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.523076923076923,
|
| 1243 |
+
"eval_loss": 0.8907297849655151,
|
| 1244 |
+
"eval_runtime": 1.1681,
|
| 1245 |
+
"eval_samples_per_second": 97.593,
|
| 1246 |
+
"eval_steps_per_second": 5.136,
|
| 1247 |
+
"step": 820
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 2.5538461538461537,
|
| 1251 |
+
"grad_norm": 1.7706941366195679,
|
| 1252 |
+
"learning_rate": 1.8564750583869374e-05,
|
| 1253 |
+
"loss": 0.2278,
|
| 1254 |
+
"step": 830
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 2.5538461538461537,
|
| 1258 |
+
"eval_loss": 0.8588199019432068,
|
| 1259 |
+
"eval_runtime": 1.1709,
|
| 1260 |
+
"eval_samples_per_second": 97.36,
|
| 1261 |
+
"eval_steps_per_second": 5.124,
|
| 1262 |
+
"step": 830
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.5846153846153848,
|
| 1266 |
+
"grad_norm": 1.4446183443069458,
|
| 1267 |
+
"learning_rate": 1.850881649528625e-05,
|
| 1268 |
+
"loss": 0.2033,
|
| 1269 |
+
"step": 840
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.5846153846153848,
|
| 1273 |
+
"eval_loss": 0.8720155954360962,
|
| 1274 |
+
"eval_runtime": 1.1716,
|
| 1275 |
+
"eval_samples_per_second": 97.304,
|
| 1276 |
+
"eval_steps_per_second": 5.121,
|
| 1277 |
+
"step": 840
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.6153846153846154,
|
| 1281 |
+
"grad_norm": 1.5155885219573975,
|
| 1282 |
+
"learning_rate": 1.845190085543795e-05,
|
| 1283 |
+
"loss": 0.2211,
|
| 1284 |
+
"step": 850
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6153846153846154,
|
| 1288 |
+
"eval_loss": 0.8732415437698364,
|
| 1289 |
+
"eval_runtime": 1.1708,
|
| 1290 |
+
"eval_samples_per_second": 97.371,
|
| 1291 |
+
"eval_steps_per_second": 5.125,
|
| 1292 |
+
"step": 850
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.646153846153846,
|
| 1296 |
+
"grad_norm": 2.934957981109619,
|
| 1297 |
+
"learning_rate": 1.839401022994006e-05,
|
| 1298 |
+
"loss": 0.21,
|
| 1299 |
+
"step": 860
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.646153846153846,
|
| 1303 |
+
"eval_loss": 0.8790932297706604,
|
| 1304 |
+
"eval_runtime": 1.1671,
|
| 1305 |
+
"eval_samples_per_second": 97.68,
|
| 1306 |
+
"eval_steps_per_second": 5.141,
|
| 1307 |
+
"step": 860
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.676923076923077,
|
| 1311 |
+
"grad_norm": 2.930457353591919,
|
| 1312 |
+
"learning_rate": 1.8335151296879576e-05,
|
| 1313 |
+
"loss": 0.2313,
|
| 1314 |
+
"step": 870
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.676923076923077,
|
| 1318 |
+
"eval_loss": 0.8555310368537903,
|
| 1319 |
+
"eval_runtime": 1.1718,
|
| 1320 |
+
"eval_samples_per_second": 97.284,
|
| 1321 |
+
"eval_steps_per_second": 5.12,
|
| 1322 |
+
"step": 870
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.707692307692308,
|
| 1326 |
+
"grad_norm": 2.029303789138794,
|
| 1327 |
+
"learning_rate": 1.82753308460445e-05,
|
| 1328 |
+
"loss": 0.2285,
|
| 1329 |
+
"step": 880
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.707692307692308,
|
| 1333 |
+
"eval_loss": 0.870873749256134,
|
| 1334 |
+
"eval_runtime": 1.1696,
|
| 1335 |
+
"eval_samples_per_second": 97.47,
|
| 1336 |
+
"eval_steps_per_second": 5.13,
|
| 1337 |
+
"step": 880
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 2.7384615384615385,
|
| 1341 |
+
"grad_norm": 1.949743390083313,
|
| 1342 |
+
"learning_rate": 1.821455577814062e-05,
|
| 1343 |
+
"loss": 0.2134,
|
| 1344 |
+
"step": 890
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.7384615384615385,
|
| 1348 |
+
"eval_loss": 0.8873838186264038,
|
| 1349 |
+
"eval_runtime": 1.1688,
|
| 1350 |
+
"eval_samples_per_second": 97.538,
|
| 1351 |
+
"eval_steps_per_second": 5.134,
|
| 1352 |
+
"step": 890
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.769230769230769,
|
| 1356 |
+
"grad_norm": 1.6945637464523315,
|
| 1357 |
+
"learning_rate": 1.8152833103995443e-05,
|
| 1358 |
+
"loss": 0.218,
|
| 1359 |
+
"step": 900
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.769230769230769,
|
| 1363 |
+
"eval_loss": 0.8798539042472839,
|
| 1364 |
+
"eval_runtime": 1.1699,
|
| 1365 |
+
"eval_samples_per_second": 97.444,
|
| 1366 |
+
"eval_steps_per_second": 5.129,
|
| 1367 |
+
"step": 900
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8,
|
| 1371 |
+
"grad_norm": 2.2784035205841064,
|
| 1372 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 1373 |
+
"loss": 0.2426,
|
| 1374 |
+
"step": 910
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8,
|
| 1378 |
+
"eval_loss": 0.8753871321678162,
|
| 1379 |
+
"eval_runtime": 1.1723,
|
| 1380 |
+
"eval_samples_per_second": 97.246,
|
| 1381 |
+
"eval_steps_per_second": 5.118,
|
| 1382 |
+
"step": 910
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.830769230769231,
|
| 1386 |
+
"grad_norm": 1.750718593597412,
|
| 1387 |
+
"learning_rate": 1.802657352603483e-05,
|
| 1388 |
+
"loss": 0.2059,
|
| 1389 |
+
"step": 920
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.830769230769231,
|
| 1393 |
+
"eval_loss": 0.8761821985244751,
|
| 1394 |
+
"eval_runtime": 1.1762,
|
| 1395 |
+
"eval_samples_per_second": 96.92,
|
| 1396 |
+
"eval_steps_per_second": 5.101,
|
| 1397 |
+
"step": 920
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.8615384615384616,
|
| 1401 |
+
"grad_norm": 1.2204371690750122,
|
| 1402 |
+
"learning_rate": 1.7962051187141377e-05,
|
| 1403 |
+
"loss": 0.2229,
|
| 1404 |
+
"step": 930
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.8615384615384616,
|
| 1408 |
+
"eval_loss": 0.8772630095481873,
|
| 1409 |
+
"eval_runtime": 1.1697,
|
| 1410 |
+
"eval_samples_per_second": 97.458,
|
| 1411 |
+
"eval_steps_per_second": 5.129,
|
| 1412 |
+
"step": 930
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.8923076923076922,
|
| 1416 |
+
"grad_norm": 2.0748281478881836,
|
| 1417 |
+
"learning_rate": 1.7896610370170452e-05,
|
| 1418 |
+
"loss": 0.2582,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.8923076923076922,
|
| 1423 |
+
"eval_loss": 0.884286105632782,
|
| 1424 |
+
"eval_runtime": 1.1773,
|
| 1425 |
+
"eval_samples_per_second": 96.829,
|
| 1426 |
+
"eval_steps_per_second": 5.096,
|
| 1427 |
+
"step": 940
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.9230769230769234,
|
| 1431 |
+
"grad_norm": 1.9051027297973633,
|
| 1432 |
+
"learning_rate": 1.7830258624176224e-05,
|
| 1433 |
+
"loss": 0.2102,
|
| 1434 |
+
"step": 950
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.9230769230769234,
|
| 1438 |
+
"eval_loss": 0.878767192363739,
|
| 1439 |
+
"eval_runtime": 1.1684,
|
| 1440 |
+
"eval_samples_per_second": 97.571,
|
| 1441 |
+
"eval_steps_per_second": 5.135,
|
| 1442 |
+
"step": 950
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.953846153846154,
|
| 1446 |
+
"grad_norm": 1.8921849727630615,
|
| 1447 |
+
"learning_rate": 1.776300360329488e-05,
|
| 1448 |
+
"loss": 0.2331,
|
| 1449 |
+
"step": 960
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.953846153846154,
|
| 1453 |
+
"eval_loss": 0.8681402206420898,
|
| 1454 |
+
"eval_runtime": 1.1693,
|
| 1455 |
+
"eval_samples_per_second": 97.497,
|
| 1456 |
+
"eval_steps_per_second": 5.131,
|
| 1457 |
+
"step": 960
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.9846153846153847,
|
| 1461 |
+
"grad_norm": 1.5593761205673218,
|
| 1462 |
+
"learning_rate": 1.769485306586166e-05,
|
| 1463 |
+
"loss": 0.2226,
|
| 1464 |
+
"step": 970
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.9846153846153847,
|
| 1468 |
+
"eval_loss": 0.8726215958595276,
|
| 1469 |
+
"eval_runtime": 1.1712,
|
| 1470 |
+
"eval_samples_per_second": 97.336,
|
| 1471 |
+
"eval_steps_per_second": 5.123,
|
| 1472 |
+
"step": 970
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.0153846153846153,
|
| 1476 |
+
"grad_norm": 1.4075913429260254,
|
| 1477 |
+
"learning_rate": 1.762581487351587e-05,
|
| 1478 |
+
"loss": 0.1535,
|
| 1479 |
+
"step": 980
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.0153846153846153,
|
| 1483 |
+
"eval_loss": 0.9113463163375854,
|
| 1484 |
+
"eval_runtime": 1.1794,
|
| 1485 |
+
"eval_samples_per_second": 96.661,
|
| 1486 |
+
"eval_steps_per_second": 5.087,
|
| 1487 |
+
"step": 980
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.046153846153846,
|
| 1491 |
+
"grad_norm": 1.410906434059143,
|
| 1492 |
+
"learning_rate": 1.7555896990294003e-05,
|
| 1493 |
+
"loss": 0.1021,
|
| 1494 |
+
"step": 990
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.046153846153846,
|
| 1498 |
+
"eval_loss": 0.9633604288101196,
|
| 1499 |
+
"eval_runtime": 1.1702,
|
| 1500 |
+
"eval_samples_per_second": 97.416,
|
| 1501 |
+
"eval_steps_per_second": 5.127,
|
| 1502 |
+
"step": 990
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 3.076923076923077,
|
| 1506 |
+
"grad_norm": 1.0470328330993652,
|
| 1507 |
+
"learning_rate": 1.7485107481711014e-05,
|
| 1508 |
+
"loss": 0.1118,
|
| 1509 |
+
"step": 1000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 3.076923076923077,
|
| 1513 |
+
"eval_loss": 0.9647616744041443,
|
| 1514 |
+
"eval_runtime": 1.1683,
|
| 1515 |
+
"eval_samples_per_second": 97.574,
|
| 1516 |
+
"eval_steps_per_second": 5.135,
|
| 1517 |
+
"step": 1000
|
| 1518 |
+
}
|
| 1519 |
+
],
|
| 1520 |
+
"logging_steps": 10,
|
| 1521 |
+
"max_steps": 3250,
|
| 1522 |
+
"num_input_tokens_seen": 0,
|
| 1523 |
+
"num_train_epochs": 10,
|
| 1524 |
+
"save_steps": 50,
|
| 1525 |
+
"total_flos": 6.244958959199846e+16,
|
| 1526 |
+
"train_batch_size": 2,
|
| 1527 |
+
"trial_name": null,
|
| 1528 |
+
"trial_params": null
|
| 1529 |
+
}
|
{checkpoint-100 → checkpoint-1000}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/added_tokens.json
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/config.json
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/generation_config.json
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/model.safetensors.index.json
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/special_tokens_map.json
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/tokenizer.json
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/tokenizer.model
RENAMED
|
File without changes
|
{checkpoint-150 → checkpoint-1050}/tokenizer_config.json
RENAMED
|
File without changes
|
checkpoint-1050/trainer_state.json
ADDED
|
@@ -0,0 +1,1604 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.230769230769231,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1050,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006153846153846154,
|
| 13 |
+
"eval_loss": 1.487821340560913,
|
| 14 |
+
"eval_runtime": 1.2374,
|
| 15 |
+
"eval_samples_per_second": 92.131,
|
| 16 |
+
"eval_steps_per_second": 4.849,
|
| 17 |
+
"step": 2
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03076923076923077,
|
| 21 |
+
"grad_norm": 20.007999420166016,
|
| 22 |
+
"learning_rate": 6.153846153846155e-07,
|
| 23 |
+
"loss": 1.4703,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03076923076923077,
|
| 28 |
+
"eval_loss": 1.4221965074539185,
|
| 29 |
+
"eval_runtime": 1.1701,
|
| 30 |
+
"eval_samples_per_second": 97.431,
|
| 31 |
+
"eval_steps_per_second": 5.128,
|
| 32 |
+
"step": 10
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06153846153846154,
|
| 36 |
+
"grad_norm": 29.657033920288086,
|
| 37 |
+
"learning_rate": 1.230769230769231e-06,
|
| 38 |
+
"loss": 1.4402,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06153846153846154,
|
| 43 |
+
"eval_loss": 1.238189935684204,
|
| 44 |
+
"eval_runtime": 1.1729,
|
| 45 |
+
"eval_samples_per_second": 97.195,
|
| 46 |
+
"eval_steps_per_second": 5.116,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.09230769230769231,
|
| 51 |
+
"grad_norm": 10.80688190460205,
|
| 52 |
+
"learning_rate": 1.8461538461538465e-06,
|
| 53 |
+
"loss": 1.0957,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09230769230769231,
|
| 58 |
+
"eval_loss": 0.908875048160553,
|
| 59 |
+
"eval_runtime": 1.1739,
|
| 60 |
+
"eval_samples_per_second": 97.111,
|
| 61 |
+
"eval_steps_per_second": 5.111,
|
| 62 |
+
"step": 30
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.12307692307692308,
|
| 66 |
+
"grad_norm": 12.914667129516602,
|
| 67 |
+
"learning_rate": 2.461538461538462e-06,
|
| 68 |
+
"loss": 0.8136,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.12307692307692308,
|
| 73 |
+
"eval_loss": 0.7436326146125793,
|
| 74 |
+
"eval_runtime": 1.1712,
|
| 75 |
+
"eval_samples_per_second": 97.34,
|
| 76 |
+
"eval_steps_per_second": 5.123,
|
| 77 |
+
"step": 40
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.15384615384615385,
|
| 81 |
+
"grad_norm": 5.206438064575195,
|
| 82 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 83 |
+
"loss": 0.7864,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.15384615384615385,
|
| 88 |
+
"eval_loss": 0.6924600601196289,
|
| 89 |
+
"eval_runtime": 1.1718,
|
| 90 |
+
"eval_samples_per_second": 97.287,
|
| 91 |
+
"eval_steps_per_second": 5.12,
|
| 92 |
+
"step": 50
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.18461538461538463,
|
| 96 |
+
"grad_norm": 5.670599460601807,
|
| 97 |
+
"learning_rate": 3.692307692307693e-06,
|
| 98 |
+
"loss": 0.622,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.18461538461538463,
|
| 103 |
+
"eval_loss": 0.6791077852249146,
|
| 104 |
+
"eval_runtime": 1.1678,
|
| 105 |
+
"eval_samples_per_second": 97.62,
|
| 106 |
+
"eval_steps_per_second": 5.138,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.2153846153846154,
|
| 111 |
+
"grad_norm": 3.6266868114471436,
|
| 112 |
+
"learning_rate": 4.307692307692308e-06,
|
| 113 |
+
"loss": 0.6672,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2153846153846154,
|
| 118 |
+
"eval_loss": 0.6631948947906494,
|
| 119 |
+
"eval_runtime": 1.1626,
|
| 120 |
+
"eval_samples_per_second": 98.06,
|
| 121 |
+
"eval_steps_per_second": 5.161,
|
| 122 |
+
"step": 70
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.24615384615384617,
|
| 126 |
+
"grad_norm": 3.1027884483337402,
|
| 127 |
+
"learning_rate": 4.923076923076924e-06,
|
| 128 |
+
"loss": 0.6097,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.24615384615384617,
|
| 133 |
+
"eval_loss": 0.6625475883483887,
|
| 134 |
+
"eval_runtime": 1.1717,
|
| 135 |
+
"eval_samples_per_second": 97.294,
|
| 136 |
+
"eval_steps_per_second": 5.121,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.27692307692307694,
|
| 141 |
+
"grad_norm": 3.233839511871338,
|
| 142 |
+
"learning_rate": 5.538461538461539e-06,
|
| 143 |
+
"loss": 0.6957,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.27692307692307694,
|
| 148 |
+
"eval_loss": 0.6590337157249451,
|
| 149 |
+
"eval_runtime": 1.1664,
|
| 150 |
+
"eval_samples_per_second": 97.738,
|
| 151 |
+
"eval_steps_per_second": 5.144,
|
| 152 |
+
"step": 90
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.3076923076923077,
|
| 156 |
+
"grad_norm": 4.932614803314209,
|
| 157 |
+
"learning_rate": 6.153846153846155e-06,
|
| 158 |
+
"loss": 0.6994,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.3076923076923077,
|
| 163 |
+
"eval_loss": 0.6592673659324646,
|
| 164 |
+
"eval_runtime": 1.1692,
|
| 165 |
+
"eval_samples_per_second": 97.502,
|
| 166 |
+
"eval_steps_per_second": 5.132,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.3384615384615385,
|
| 171 |
+
"grad_norm": 2.642303466796875,
|
| 172 |
+
"learning_rate": 6.76923076923077e-06,
|
| 173 |
+
"loss": 0.6501,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.3384615384615385,
|
| 178 |
+
"eval_loss": 0.6589992642402649,
|
| 179 |
+
"eval_runtime": 1.1669,
|
| 180 |
+
"eval_samples_per_second": 97.691,
|
| 181 |
+
"eval_steps_per_second": 5.142,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.36923076923076925,
|
| 186 |
+
"grad_norm": 3.45409893989563,
|
| 187 |
+
"learning_rate": 7.384615384615386e-06,
|
| 188 |
+
"loss": 0.7512,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.36923076923076925,
|
| 193 |
+
"eval_loss": 0.6568591594696045,
|
| 194 |
+
"eval_runtime": 1.1778,
|
| 195 |
+
"eval_samples_per_second": 96.79,
|
| 196 |
+
"eval_steps_per_second": 5.094,
|
| 197 |
+
"step": 120
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.4,
|
| 201 |
+
"grad_norm": 8.460351943969727,
|
| 202 |
+
"learning_rate": 8.000000000000001e-06,
|
| 203 |
+
"loss": 0.5996,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4,
|
| 208 |
+
"eval_loss": 0.6578297019004822,
|
| 209 |
+
"eval_runtime": 1.1683,
|
| 210 |
+
"eval_samples_per_second": 97.575,
|
| 211 |
+
"eval_steps_per_second": 5.136,
|
| 212 |
+
"step": 130
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4307692307692308,
|
| 216 |
+
"grad_norm": 2.4495465755462646,
|
| 217 |
+
"learning_rate": 8.615384615384617e-06,
|
| 218 |
+
"loss": 0.6498,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4307692307692308,
|
| 223 |
+
"eval_loss": 0.6588199138641357,
|
| 224 |
+
"eval_runtime": 1.1664,
|
| 225 |
+
"eval_samples_per_second": 97.739,
|
| 226 |
+
"eval_steps_per_second": 5.144,
|
| 227 |
+
"step": 140
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.46153846153846156,
|
| 231 |
+
"grad_norm": 2.6779162883758545,
|
| 232 |
+
"learning_rate": 9.230769230769232e-06,
|
| 233 |
+
"loss": 0.703,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.46153846153846156,
|
| 238 |
+
"eval_loss": 0.6626676321029663,
|
| 239 |
+
"eval_runtime": 1.1764,
|
| 240 |
+
"eval_samples_per_second": 96.904,
|
| 241 |
+
"eval_steps_per_second": 5.1,
|
| 242 |
+
"step": 150
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.49230769230769234,
|
| 246 |
+
"grad_norm": 8.147880554199219,
|
| 247 |
+
"learning_rate": 9.846153846153848e-06,
|
| 248 |
+
"loss": 0.6649,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.49230769230769234,
|
| 253 |
+
"eval_loss": 0.6644802689552307,
|
| 254 |
+
"eval_runtime": 1.1701,
|
| 255 |
+
"eval_samples_per_second": 97.424,
|
| 256 |
+
"eval_steps_per_second": 5.128,
|
| 257 |
+
"step": 160
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.5230769230769231,
|
| 261 |
+
"grad_norm": 2.7410728931427,
|
| 262 |
+
"learning_rate": 1.0461538461538463e-05,
|
| 263 |
+
"loss": 0.6704,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.5230769230769231,
|
| 268 |
+
"eval_loss": 0.6591455340385437,
|
| 269 |
+
"eval_runtime": 1.1722,
|
| 270 |
+
"eval_samples_per_second": 97.253,
|
| 271 |
+
"eval_steps_per_second": 5.119,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.5538461538461539,
|
| 276 |
+
"grad_norm": 6.886598587036133,
|
| 277 |
+
"learning_rate": 1.1076923076923079e-05,
|
| 278 |
+
"loss": 0.6073,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.5538461538461539,
|
| 283 |
+
"eval_loss": 0.6617754697799683,
|
| 284 |
+
"eval_runtime": 1.1682,
|
| 285 |
+
"eval_samples_per_second": 97.588,
|
| 286 |
+
"eval_steps_per_second": 5.136,
|
| 287 |
+
"step": 180
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.5846153846153846,
|
| 291 |
+
"grad_norm": 4.26249361038208,
|
| 292 |
+
"learning_rate": 1.1692307692307694e-05,
|
| 293 |
+
"loss": 0.7556,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.5846153846153846,
|
| 298 |
+
"eval_loss": 0.6659030318260193,
|
| 299 |
+
"eval_runtime": 1.1721,
|
| 300 |
+
"eval_samples_per_second": 97.265,
|
| 301 |
+
"eval_steps_per_second": 5.119,
|
| 302 |
+
"step": 190
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.6153846153846154,
|
| 306 |
+
"grad_norm": 4.204222202301025,
|
| 307 |
+
"learning_rate": 1.230769230769231e-05,
|
| 308 |
+
"loss": 0.6583,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.6153846153846154,
|
| 313 |
+
"eval_loss": 0.6660778522491455,
|
| 314 |
+
"eval_runtime": 1.1728,
|
| 315 |
+
"eval_samples_per_second": 97.201,
|
| 316 |
+
"eval_steps_per_second": 5.116,
|
| 317 |
+
"step": 200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6461538461538462,
|
| 321 |
+
"grad_norm": 3.154658079147339,
|
| 322 |
+
"learning_rate": 1.2923076923076925e-05,
|
| 323 |
+
"loss": 0.7288,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6461538461538462,
|
| 328 |
+
"eval_loss": 0.6642422676086426,
|
| 329 |
+
"eval_runtime": 1.166,
|
| 330 |
+
"eval_samples_per_second": 97.772,
|
| 331 |
+
"eval_steps_per_second": 5.146,
|
| 332 |
+
"step": 210
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.676923076923077,
|
| 336 |
+
"grad_norm": 3.248359203338623,
|
| 337 |
+
"learning_rate": 1.353846153846154e-05,
|
| 338 |
+
"loss": 0.6702,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.676923076923077,
|
| 343 |
+
"eval_loss": 0.6680281758308411,
|
| 344 |
+
"eval_runtime": 1.1691,
|
| 345 |
+
"eval_samples_per_second": 97.507,
|
| 346 |
+
"eval_steps_per_second": 5.132,
|
| 347 |
+
"step": 220
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.7076923076923077,
|
| 351 |
+
"grad_norm": 2.5507216453552246,
|
| 352 |
+
"learning_rate": 1.4153846153846156e-05,
|
| 353 |
+
"loss": 0.66,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.7076923076923077,
|
| 358 |
+
"eval_loss": 0.6713310480117798,
|
| 359 |
+
"eval_runtime": 1.167,
|
| 360 |
+
"eval_samples_per_second": 97.683,
|
| 361 |
+
"eval_steps_per_second": 5.141,
|
| 362 |
+
"step": 230
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.7384615384615385,
|
| 366 |
+
"grad_norm": 2.1180007457733154,
|
| 367 |
+
"learning_rate": 1.4769230769230772e-05,
|
| 368 |
+
"loss": 0.6172,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.7384615384615385,
|
| 373 |
+
"eval_loss": 0.6759226322174072,
|
| 374 |
+
"eval_runtime": 1.1758,
|
| 375 |
+
"eval_samples_per_second": 96.954,
|
| 376 |
+
"eval_steps_per_second": 5.103,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.7692307692307693,
|
| 381 |
+
"grad_norm": 2.8318371772766113,
|
| 382 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 383 |
+
"loss": 0.5954,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.7692307692307693,
|
| 388 |
+
"eval_loss": 0.6854314208030701,
|
| 389 |
+
"eval_runtime": 1.1698,
|
| 390 |
+
"eval_samples_per_second": 97.457,
|
| 391 |
+
"eval_steps_per_second": 5.129,
|
| 392 |
+
"step": 250
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.8,
|
| 396 |
+
"grad_norm": 3.519829034805298,
|
| 397 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 398 |
+
"loss": 0.6639,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.8,
|
| 403 |
+
"eval_loss": 0.6849851012229919,
|
| 404 |
+
"eval_runtime": 1.1732,
|
| 405 |
+
"eval_samples_per_second": 97.173,
|
| 406 |
+
"eval_steps_per_second": 5.114,
|
| 407 |
+
"step": 260
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.8307692307692308,
|
| 411 |
+
"grad_norm": 3.7909388542175293,
|
| 412 |
+
"learning_rate": 1.6615384615384618e-05,
|
| 413 |
+
"loss": 0.6892,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.8307692307692308,
|
| 418 |
+
"eval_loss": 0.6863855123519897,
|
| 419 |
+
"eval_runtime": 1.1689,
|
| 420 |
+
"eval_samples_per_second": 97.53,
|
| 421 |
+
"eval_steps_per_second": 5.133,
|
| 422 |
+
"step": 270
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8615384615384616,
|
| 426 |
+
"grad_norm": 8.278326988220215,
|
| 427 |
+
"learning_rate": 1.7230769230769234e-05,
|
| 428 |
+
"loss": 0.6014,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8615384615384616,
|
| 433 |
+
"eval_loss": 0.6890373826026917,
|
| 434 |
+
"eval_runtime": 1.1722,
|
| 435 |
+
"eval_samples_per_second": 97.251,
|
| 436 |
+
"eval_steps_per_second": 5.118,
|
| 437 |
+
"step": 280
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8923076923076924,
|
| 441 |
+
"grad_norm": 2.616001605987549,
|
| 442 |
+
"learning_rate": 1.784615384615385e-05,
|
| 443 |
+
"loss": 0.6775,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.8923076923076924,
|
| 448 |
+
"eval_loss": 0.6797110438346863,
|
| 449 |
+
"eval_runtime": 1.1888,
|
| 450 |
+
"eval_samples_per_second": 95.897,
|
| 451 |
+
"eval_steps_per_second": 5.047,
|
| 452 |
+
"step": 290
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.9230769230769231,
|
| 456 |
+
"grad_norm": 2.4450528621673584,
|
| 457 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 458 |
+
"loss": 0.7434,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.9230769230769231,
|
| 463 |
+
"eval_loss": 0.6779448986053467,
|
| 464 |
+
"eval_runtime": 1.1722,
|
| 465 |
+
"eval_samples_per_second": 97.254,
|
| 466 |
+
"eval_steps_per_second": 5.119,
|
| 467 |
+
"step": 300
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.9538461538461539,
|
| 471 |
+
"grad_norm": 4.121164798736572,
|
| 472 |
+
"learning_rate": 1.907692307692308e-05,
|
| 473 |
+
"loss": 0.6413,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.9538461538461539,
|
| 478 |
+
"eval_loss": 0.6819297671318054,
|
| 479 |
+
"eval_runtime": 1.1683,
|
| 480 |
+
"eval_samples_per_second": 97.574,
|
| 481 |
+
"eval_steps_per_second": 5.135,
|
| 482 |
+
"step": 310
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.9846153846153847,
|
| 486 |
+
"grad_norm": 2.057229995727539,
|
| 487 |
+
"learning_rate": 1.9692307692307696e-05,
|
| 488 |
+
"loss": 0.6511,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.9846153846153847,
|
| 493 |
+
"eval_loss": 0.6861481666564941,
|
| 494 |
+
"eval_runtime": 1.1699,
|
| 495 |
+
"eval_samples_per_second": 97.444,
|
| 496 |
+
"eval_steps_per_second": 5.129,
|
| 497 |
+
"step": 320
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.0153846153846153,
|
| 501 |
+
"grad_norm": 1.7295210361480713,
|
| 502 |
+
"learning_rate": 1.9999855802751384e-05,
|
| 503 |
+
"loss": 0.5872,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.0153846153846153,
|
| 508 |
+
"eval_loss": 0.7481781244277954,
|
| 509 |
+
"eval_runtime": 1.172,
|
| 510 |
+
"eval_samples_per_second": 97.273,
|
| 511 |
+
"eval_steps_per_second": 5.12,
|
| 512 |
+
"step": 330
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.0461538461538462,
|
| 516 |
+
"grad_norm": 3.6910336017608643,
|
| 517 |
+
"learning_rate": 1.9998702249713747e-05,
|
| 518 |
+
"loss": 0.438,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.0461538461538462,
|
| 523 |
+
"eval_loss": 0.7471169829368591,
|
| 524 |
+
"eval_runtime": 1.175,
|
| 525 |
+
"eval_samples_per_second": 97.019,
|
| 526 |
+
"eval_steps_per_second": 5.106,
|
| 527 |
+
"step": 340
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.0769230769230769,
|
| 531 |
+
"grad_norm": 3.9220831394195557,
|
| 532 |
+
"learning_rate": 1.9996395276708856e-05,
|
| 533 |
+
"loss": 0.4141,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0769230769230769,
|
| 538 |
+
"eval_loss": 0.7425897121429443,
|
| 539 |
+
"eval_runtime": 1.1888,
|
| 540 |
+
"eval_samples_per_second": 95.893,
|
| 541 |
+
"eval_steps_per_second": 5.047,
|
| 542 |
+
"step": 350
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.1076923076923078,
|
| 546 |
+
"grad_norm": 2.1136879920959473,
|
| 547 |
+
"learning_rate": 1.9992935149862116e-05,
|
| 548 |
+
"loss": 0.4173,
|
| 549 |
+
"step": 360
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1076923076923078,
|
| 553 |
+
"eval_loss": 0.7477542161941528,
|
| 554 |
+
"eval_runtime": 1.1705,
|
| 555 |
+
"eval_samples_per_second": 97.398,
|
| 556 |
+
"eval_steps_per_second": 5.126,
|
| 557 |
+
"step": 360
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.1384615384615384,
|
| 561 |
+
"grad_norm": 2.187299966812134,
|
| 562 |
+
"learning_rate": 1.998832226832327e-05,
|
| 563 |
+
"loss": 0.4242,
|
| 564 |
+
"step": 370
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.1384615384615384,
|
| 568 |
+
"eval_loss": 0.7475994229316711,
|
| 569 |
+
"eval_runtime": 1.1683,
|
| 570 |
+
"eval_samples_per_second": 97.582,
|
| 571 |
+
"eval_steps_per_second": 5.136,
|
| 572 |
+
"step": 370
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.1692307692307693,
|
| 576 |
+
"grad_norm": 2.683750629425049,
|
| 577 |
+
"learning_rate": 1.9982557164220335e-05,
|
| 578 |
+
"loss": 0.4413,
|
| 579 |
+
"step": 380
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.1692307692307693,
|
| 583 |
+
"eval_loss": 0.7552060484886169,
|
| 584 |
+
"eval_runtime": 1.1689,
|
| 585 |
+
"eval_samples_per_second": 97.529,
|
| 586 |
+
"eval_steps_per_second": 5.133,
|
| 587 |
+
"step": 380
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.2,
|
| 591 |
+
"grad_norm": 3.312657117843628,
|
| 592 |
+
"learning_rate": 1.9975640502598243e-05,
|
| 593 |
+
"loss": 0.3361,
|
| 594 |
+
"step": 390
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.2,
|
| 598 |
+
"eval_loss": 0.7630029916763306,
|
| 599 |
+
"eval_runtime": 1.1689,
|
| 600 |
+
"eval_samples_per_second": 97.529,
|
| 601 |
+
"eval_steps_per_second": 5.133,
|
| 602 |
+
"step": 390
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.2307692307692308,
|
| 606 |
+
"grad_norm": 2.3290536403656006,
|
| 607 |
+
"learning_rate": 1.9967573081342103e-05,
|
| 608 |
+
"loss": 0.4174,
|
| 609 |
+
"step": 400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.2307692307692308,
|
| 613 |
+
"eval_loss": 0.7669293880462646,
|
| 614 |
+
"eval_runtime": 1.1711,
|
| 615 |
+
"eval_samples_per_second": 97.345,
|
| 616 |
+
"eval_steps_per_second": 5.123,
|
| 617 |
+
"step": 400
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.2615384615384615,
|
| 621 |
+
"grad_norm": 2.4483697414398193,
|
| 622 |
+
"learning_rate": 1.9958355831085155e-05,
|
| 623 |
+
"loss": 0.4045,
|
| 624 |
+
"step": 410
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.2615384615384615,
|
| 628 |
+
"eval_loss": 0.7652024626731873,
|
| 629 |
+
"eval_runtime": 1.1697,
|
| 630 |
+
"eval_samples_per_second": 97.461,
|
| 631 |
+
"eval_steps_per_second": 5.13,
|
| 632 |
+
"step": 410
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.2923076923076924,
|
| 636 |
+
"grad_norm": 1.703616976737976,
|
| 637 |
+
"learning_rate": 1.9947989815101444e-05,
|
| 638 |
+
"loss": 0.4292,
|
| 639 |
+
"step": 420
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.2923076923076924,
|
| 643 |
+
"eval_loss": 0.7553891539573669,
|
| 644 |
+
"eval_runtime": 1.1719,
|
| 645 |
+
"eval_samples_per_second": 97.277,
|
| 646 |
+
"eval_steps_per_second": 5.12,
|
| 647 |
+
"step": 420
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.323076923076923,
|
| 651 |
+
"grad_norm": 1.6245399713516235,
|
| 652 |
+
"learning_rate": 1.9936476229183133e-05,
|
| 653 |
+
"loss": 0.4164,
|
| 654 |
+
"step": 430
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.323076923076923,
|
| 658 |
+
"eval_loss": 0.7711736559867859,
|
| 659 |
+
"eval_runtime": 1.1776,
|
| 660 |
+
"eval_samples_per_second": 96.806,
|
| 661 |
+
"eval_steps_per_second": 5.095,
|
| 662 |
+
"step": 430
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.353846153846154,
|
| 666 |
+
"grad_norm": 1.9511611461639404,
|
| 667 |
+
"learning_rate": 1.992381640150257e-05,
|
| 668 |
+
"loss": 0.4337,
|
| 669 |
+
"step": 440
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.353846153846154,
|
| 673 |
+
"eval_loss": 0.7438372373580933,
|
| 674 |
+
"eval_runtime": 1.1692,
|
| 675 |
+
"eval_samples_per_second": 97.5,
|
| 676 |
+
"eval_steps_per_second": 5.132,
|
| 677 |
+
"step": 440
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.3846153846153846,
|
| 681 |
+
"grad_norm": 2.466737747192383,
|
| 682 |
+
"learning_rate": 1.9910011792459086e-05,
|
| 683 |
+
"loss": 0.3712,
|
| 684 |
+
"step": 450
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.3846153846153846,
|
| 688 |
+
"eval_loss": 0.8045614361763,
|
| 689 |
+
"eval_runtime": 1.1689,
|
| 690 |
+
"eval_samples_per_second": 97.53,
|
| 691 |
+
"eval_steps_per_second": 5.133,
|
| 692 |
+
"step": 450
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.4153846153846155,
|
| 696 |
+
"grad_norm": 2.473851203918457,
|
| 697 |
+
"learning_rate": 1.9895063994510512e-05,
|
| 698 |
+
"loss": 0.516,
|
| 699 |
+
"step": 460
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.4153846153846155,
|
| 703 |
+
"eval_loss": 0.7463352084159851,
|
| 704 |
+
"eval_runtime": 1.1697,
|
| 705 |
+
"eval_samples_per_second": 97.464,
|
| 706 |
+
"eval_steps_per_second": 5.13,
|
| 707 |
+
"step": 460
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.4461538461538461,
|
| 711 |
+
"grad_norm": 2.757955312728882,
|
| 712 |
+
"learning_rate": 1.9878974731989487e-05,
|
| 713 |
+
"loss": 0.4965,
|
| 714 |
+
"step": 470
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.4461538461538461,
|
| 718 |
+
"eval_loss": 0.7571713924407959,
|
| 719 |
+
"eval_runtime": 1.1688,
|
| 720 |
+
"eval_samples_per_second": 97.534,
|
| 721 |
+
"eval_steps_per_second": 5.133,
|
| 722 |
+
"step": 470
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.476923076923077,
|
| 726 |
+
"grad_norm": 3.9884915351867676,
|
| 727 |
+
"learning_rate": 1.9861745860904538e-05,
|
| 728 |
+
"loss": 0.3973,
|
| 729 |
+
"step": 480
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.476923076923077,
|
| 733 |
+
"eval_loss": 0.7763081789016724,
|
| 734 |
+
"eval_runtime": 1.168,
|
| 735 |
+
"eval_samples_per_second": 97.605,
|
| 736 |
+
"eval_steps_per_second": 5.137,
|
| 737 |
+
"step": 480
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5076923076923077,
|
| 741 |
+
"grad_norm": 1.9474631547927856,
|
| 742 |
+
"learning_rate": 1.9843379368725978e-05,
|
| 743 |
+
"loss": 0.4625,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5076923076923077,
|
| 748 |
+
"eval_loss": 0.7595367431640625,
|
| 749 |
+
"eval_runtime": 1.1708,
|
| 750 |
+
"eval_samples_per_second": 97.37,
|
| 751 |
+
"eval_steps_per_second": 5.125,
|
| 752 |
+
"step": 490
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5384615384615383,
|
| 756 |
+
"grad_norm": 2.2907869815826416,
|
| 757 |
+
"learning_rate": 1.9823877374156647e-05,
|
| 758 |
+
"loss": 0.4371,
|
| 759 |
+
"step": 500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5384615384615383,
|
| 763 |
+
"eval_loss": 0.762865424156189,
|
| 764 |
+
"eval_runtime": 1.1766,
|
| 765 |
+
"eval_samples_per_second": 96.891,
|
| 766 |
+
"eval_steps_per_second": 5.1,
|
| 767 |
+
"step": 500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.5692307692307692,
|
| 771 |
+
"grad_norm": 3.067608594894409,
|
| 772 |
+
"learning_rate": 1.9803242126887496e-05,
|
| 773 |
+
"loss": 0.45,
|
| 774 |
+
"step": 510
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.5692307692307692,
|
| 778 |
+
"eval_loss": 0.7638718485832214,
|
| 779 |
+
"eval_runtime": 1.1684,
|
| 780 |
+
"eval_samples_per_second": 97.568,
|
| 781 |
+
"eval_steps_per_second": 5.135,
|
| 782 |
+
"step": 510
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.6,
|
| 786 |
+
"grad_norm": 2.1113197803497314,
|
| 787 |
+
"learning_rate": 1.9781476007338058e-05,
|
| 788 |
+
"loss": 0.4524,
|
| 789 |
+
"step": 520
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.6,
|
| 793 |
+
"eval_loss": 0.7560412883758545,
|
| 794 |
+
"eval_runtime": 1.1719,
|
| 795 |
+
"eval_samples_per_second": 97.278,
|
| 796 |
+
"eval_steps_per_second": 5.12,
|
| 797 |
+
"step": 520
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.6307692307692307,
|
| 801 |
+
"grad_norm": 1.889585256576538,
|
| 802 |
+
"learning_rate": 1.9758581526381878e-05,
|
| 803 |
+
"loss": 0.4287,
|
| 804 |
+
"step": 530
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.6307692307692307,
|
| 808 |
+
"eval_loss": 0.7591350674629211,
|
| 809 |
+
"eval_runtime": 1.1693,
|
| 810 |
+
"eval_samples_per_second": 97.495,
|
| 811 |
+
"eval_steps_per_second": 5.131,
|
| 812 |
+
"step": 530
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.6615384615384614,
|
| 816 |
+
"grad_norm": 2.519418716430664,
|
| 817 |
+
"learning_rate": 1.973456132505684e-05,
|
| 818 |
+
"loss": 0.4603,
|
| 819 |
+
"step": 540
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.6615384615384614,
|
| 823 |
+
"eval_loss": 0.7536805868148804,
|
| 824 |
+
"eval_runtime": 1.1799,
|
| 825 |
+
"eval_samples_per_second": 96.617,
|
| 826 |
+
"eval_steps_per_second": 5.085,
|
| 827 |
+
"step": 540
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.6923076923076923,
|
| 831 |
+
"grad_norm": 2.7455596923828125,
|
| 832 |
+
"learning_rate": 1.9709418174260523e-05,
|
| 833 |
+
"loss": 0.4293,
|
| 834 |
+
"step": 550
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.6923076923076923,
|
| 838 |
+
"eval_loss": 0.763169527053833,
|
| 839 |
+
"eval_runtime": 1.1696,
|
| 840 |
+
"eval_samples_per_second": 97.47,
|
| 841 |
+
"eval_steps_per_second": 5.13,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7230769230769232,
|
| 846 |
+
"grad_norm": 1.9823411703109741,
|
| 847 |
+
"learning_rate": 1.9683154974430544e-05,
|
| 848 |
+
"loss": 0.4643,
|
| 849 |
+
"step": 560
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7230769230769232,
|
| 853 |
+
"eval_loss": 0.7607314586639404,
|
| 854 |
+
"eval_runtime": 1.1712,
|
| 855 |
+
"eval_samples_per_second": 97.336,
|
| 856 |
+
"eval_steps_per_second": 5.123,
|
| 857 |
+
"step": 560
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.7538461538461538,
|
| 861 |
+
"grad_norm": 2.2437684535980225,
|
| 862 |
+
"learning_rate": 1.965577475520999e-05,
|
| 863 |
+
"loss": 0.452,
|
| 864 |
+
"step": 570
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.7538461538461538,
|
| 868 |
+
"eval_loss": 0.7615717649459839,
|
| 869 |
+
"eval_runtime": 1.1946,
|
| 870 |
+
"eval_samples_per_second": 95.425,
|
| 871 |
+
"eval_steps_per_second": 5.022,
|
| 872 |
+
"step": 570
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.7846153846153845,
|
| 876 |
+
"grad_norm": 3.0740182399749756,
|
| 877 |
+
"learning_rate": 1.962728067509791e-05,
|
| 878 |
+
"loss": 0.4885,
|
| 879 |
+
"step": 580
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.7846153846153845,
|
| 883 |
+
"eval_loss": 0.7764689326286316,
|
| 884 |
+
"eval_runtime": 1.1714,
|
| 885 |
+
"eval_samples_per_second": 97.32,
|
| 886 |
+
"eval_steps_per_second": 5.122,
|
| 887 |
+
"step": 580
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.8153846153846154,
|
| 891 |
+
"grad_norm": 2.050313711166382,
|
| 892 |
+
"learning_rate": 1.9597676021084962e-05,
|
| 893 |
+
"loss": 0.4992,
|
| 894 |
+
"step": 590
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.8153846153846154,
|
| 898 |
+
"eval_loss": 0.7457339763641357,
|
| 899 |
+
"eval_runtime": 1.1747,
|
| 900 |
+
"eval_samples_per_second": 97.045,
|
| 901 |
+
"eval_steps_per_second": 5.108,
|
| 902 |
+
"step": 590
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.8461538461538463,
|
| 906 |
+
"grad_norm": 2.349470853805542,
|
| 907 |
+
"learning_rate": 1.9566964208274254e-05,
|
| 908 |
+
"loss": 0.3963,
|
| 909 |
+
"step": 600
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.8461538461538463,
|
| 913 |
+
"eval_loss": 0.7712545990943909,
|
| 914 |
+
"eval_runtime": 1.17,
|
| 915 |
+
"eval_samples_per_second": 97.433,
|
| 916 |
+
"eval_steps_per_second": 5.128,
|
| 917 |
+
"step": 600
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.876923076923077,
|
| 921 |
+
"grad_norm": 1.947912573814392,
|
| 922 |
+
"learning_rate": 1.9535148779487365e-05,
|
| 923 |
+
"loss": 0.4388,
|
| 924 |
+
"step": 610
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.876923076923077,
|
| 928 |
+
"eval_loss": 0.772519052028656,
|
| 929 |
+
"eval_runtime": 1.1717,
|
| 930 |
+
"eval_samples_per_second": 97.296,
|
| 931 |
+
"eval_steps_per_second": 5.121,
|
| 932 |
+
"step": 610
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.9076923076923076,
|
| 936 |
+
"grad_norm": 1.8837225437164307,
|
| 937 |
+
"learning_rate": 1.9502233404855672e-05,
|
| 938 |
+
"loss": 0.3965,
|
| 939 |
+
"step": 620
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.9076923076923076,
|
| 943 |
+
"eval_loss": 0.7594517469406128,
|
| 944 |
+
"eval_runtime": 1.2215,
|
| 945 |
+
"eval_samples_per_second": 93.326,
|
| 946 |
+
"eval_steps_per_second": 4.912,
|
| 947 |
+
"step": 620
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9384615384615385,
|
| 951 |
+
"grad_norm": 1.8398792743682861,
|
| 952 |
+
"learning_rate": 1.946822188139696e-05,
|
| 953 |
+
"loss": 0.3898,
|
| 954 |
+
"step": 630
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.9384615384615385,
|
| 958 |
+
"eval_loss": 0.7748451828956604,
|
| 959 |
+
"eval_runtime": 1.2222,
|
| 960 |
+
"eval_samples_per_second": 93.273,
|
| 961 |
+
"eval_steps_per_second": 4.909,
|
| 962 |
+
"step": 630
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.9692307692307693,
|
| 966 |
+
"grad_norm": 2.3971219062805176,
|
| 967 |
+
"learning_rate": 1.9433118132577432e-05,
|
| 968 |
+
"loss": 0.4253,
|
| 969 |
+
"step": 640
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.9692307692307693,
|
| 973 |
+
"eval_loss": 0.7729252576828003,
|
| 974 |
+
"eval_runtime": 1.1743,
|
| 975 |
+
"eval_samples_per_second": 97.081,
|
| 976 |
+
"eval_steps_per_second": 5.11,
|
| 977 |
+
"step": 640
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.0,
|
| 981 |
+
"grad_norm": 1.657328724861145,
|
| 982 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 983 |
+
"loss": 0.4898,
|
| 984 |
+
"step": 650
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.0,
|
| 988 |
+
"eval_loss": 0.7602188587188721,
|
| 989 |
+
"eval_runtime": 1.1738,
|
| 990 |
+
"eval_samples_per_second": 97.117,
|
| 991 |
+
"eval_steps_per_second": 5.111,
|
| 992 |
+
"step": 650
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.0307692307692307,
|
| 996 |
+
"grad_norm": 1.3759773969650269,
|
| 997 |
+
"learning_rate": 1.935965028223259e-05,
|
| 998 |
+
"loss": 0.2182,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.0307692307692307,
|
| 1003 |
+
"eval_loss": 0.8499692678451538,
|
| 1004 |
+
"eval_runtime": 1.1719,
|
| 1005 |
+
"eval_samples_per_second": 97.276,
|
| 1006 |
+
"eval_steps_per_second": 5.12,
|
| 1007 |
+
"step": 660
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.0615384615384613,
|
| 1011 |
+
"grad_norm": 1.4757074117660522,
|
| 1012 |
+
"learning_rate": 1.932129465573568e-05,
|
| 1013 |
+
"loss": 0.2196,
|
| 1014 |
+
"step": 670
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.0615384615384613,
|
| 1018 |
+
"eval_loss": 0.866599977016449,
|
| 1019 |
+
"eval_runtime": 1.1739,
|
| 1020 |
+
"eval_samples_per_second": 97.115,
|
| 1021 |
+
"eval_steps_per_second": 5.111,
|
| 1022 |
+
"step": 670
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.0923076923076924,
|
| 1026 |
+
"grad_norm": 1.2500200271606445,
|
| 1027 |
+
"learning_rate": 1.9281863752957095e-05,
|
| 1028 |
+
"loss": 0.1772,
|
| 1029 |
+
"step": 680
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.0923076923076924,
|
| 1033 |
+
"eval_loss": 0.8642630577087402,
|
| 1034 |
+
"eval_runtime": 1.1753,
|
| 1035 |
+
"eval_samples_per_second": 96.999,
|
| 1036 |
+
"eval_steps_per_second": 5.105,
|
| 1037 |
+
"step": 680
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.123076923076923,
|
| 1041 |
+
"grad_norm": 2.561828851699829,
|
| 1042 |
+
"learning_rate": 1.92413621225262e-05,
|
| 1043 |
+
"loss": 0.2014,
|
| 1044 |
+
"step": 690
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.123076923076923,
|
| 1048 |
+
"eval_loss": 0.8872668743133545,
|
| 1049 |
+
"eval_runtime": 1.1761,
|
| 1050 |
+
"eval_samples_per_second": 96.933,
|
| 1051 |
+
"eval_steps_per_second": 5.102,
|
| 1052 |
+
"step": 690
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.1538461538461537,
|
| 1056 |
+
"grad_norm": 1.8431521654129028,
|
| 1057 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 1058 |
+
"loss": 0.2245,
|
| 1059 |
+
"step": 700
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.1538461538461537,
|
| 1063 |
+
"eval_loss": 0.8799911737442017,
|
| 1064 |
+
"eval_runtime": 1.1766,
|
| 1065 |
+
"eval_samples_per_second": 96.89,
|
| 1066 |
+
"eval_steps_per_second": 5.099,
|
| 1067 |
+
"step": 700
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.184615384615385,
|
| 1071 |
+
"grad_norm": 1.4718531370162964,
|
| 1072 |
+
"learning_rate": 1.915716549026541e-05,
|
| 1073 |
+
"loss": 0.1911,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.184615384615385,
|
| 1078 |
+
"eval_loss": 0.8806982636451721,
|
| 1079 |
+
"eval_runtime": 1.1696,
|
| 1080 |
+
"eval_samples_per_second": 97.473,
|
| 1081 |
+
"eval_steps_per_second": 5.13,
|
| 1082 |
+
"step": 710
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.2153846153846155,
|
| 1086 |
+
"grad_norm": 2.061520576477051,
|
| 1087 |
+
"learning_rate": 1.9113480201103658e-05,
|
| 1088 |
+
"loss": 0.232,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.2153846153846155,
|
| 1093 |
+
"eval_loss": 0.8812103271484375,
|
| 1094 |
+
"eval_runtime": 1.1676,
|
| 1095 |
+
"eval_samples_per_second": 97.636,
|
| 1096 |
+
"eval_steps_per_second": 5.139,
|
| 1097 |
+
"step": 720
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.246153846153846,
|
| 1101 |
+
"grad_norm": 2.530285596847534,
|
| 1102 |
+
"learning_rate": 1.9068743608505454e-05,
|
| 1103 |
+
"loss": 0.2232,
|
| 1104 |
+
"step": 730
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.246153846153846,
|
| 1108 |
+
"eval_loss": 0.8682798147201538,
|
| 1109 |
+
"eval_runtime": 1.1687,
|
| 1110 |
+
"eval_samples_per_second": 97.545,
|
| 1111 |
+
"eval_steps_per_second": 5.134,
|
| 1112 |
+
"step": 730
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.276923076923077,
|
| 1116 |
+
"grad_norm": 1.8433207273483276,
|
| 1117 |
+
"learning_rate": 1.902296087314845e-05,
|
| 1118 |
+
"loss": 0.2315,
|
| 1119 |
+
"step": 740
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.276923076923077,
|
| 1123 |
+
"eval_loss": 0.8747490048408508,
|
| 1124 |
+
"eval_runtime": 1.1702,
|
| 1125 |
+
"eval_samples_per_second": 97.421,
|
| 1126 |
+
"eval_steps_per_second": 5.127,
|
| 1127 |
+
"step": 740
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.3076923076923075,
|
| 1131 |
+
"grad_norm": 2.1529664993286133,
|
| 1132 |
+
"learning_rate": 1.8976137276390145e-05,
|
| 1133 |
+
"loss": 0.239,
|
| 1134 |
+
"step": 750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.3076923076923075,
|
| 1138 |
+
"eval_loss": 0.8778530359268188,
|
| 1139 |
+
"eval_runtime": 1.1674,
|
| 1140 |
+
"eval_samples_per_second": 97.656,
|
| 1141 |
+
"eval_steps_per_second": 5.14,
|
| 1142 |
+
"step": 750
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.3384615384615386,
|
| 1146 |
+
"grad_norm": 1.839979887008667,
|
| 1147 |
+
"learning_rate": 1.892827821965864e-05,
|
| 1148 |
+
"loss": 0.2228,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.3384615384615386,
|
| 1153 |
+
"eval_loss": 0.8769873976707458,
|
| 1154 |
+
"eval_runtime": 1.1718,
|
| 1155 |
+
"eval_samples_per_second": 97.288,
|
| 1156 |
+
"eval_steps_per_second": 5.12,
|
| 1157 |
+
"step": 760
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.3692307692307693,
|
| 1161 |
+
"grad_norm": 1.3814674615859985,
|
| 1162 |
+
"learning_rate": 1.8879389223829592e-05,
|
| 1163 |
+
"loss": 0.1999,
|
| 1164 |
+
"step": 770
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.3692307692307693,
|
| 1168 |
+
"eval_loss": 0.8731983304023743,
|
| 1169 |
+
"eval_runtime": 1.1723,
|
| 1170 |
+
"eval_samples_per_second": 97.242,
|
| 1171 |
+
"eval_steps_per_second": 5.118,
|
| 1172 |
+
"step": 770
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.4,
|
| 1176 |
+
"grad_norm": 1.5382564067840576,
|
| 1177 |
+
"learning_rate": 1.8829475928589272e-05,
|
| 1178 |
+
"loss": 0.1888,
|
| 1179 |
+
"step": 780
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.4,
|
| 1183 |
+
"eval_loss": 0.8794592022895813,
|
| 1184 |
+
"eval_runtime": 1.1678,
|
| 1185 |
+
"eval_samples_per_second": 97.622,
|
| 1186 |
+
"eval_steps_per_second": 5.138,
|
| 1187 |
+
"step": 780
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.430769230769231,
|
| 1191 |
+
"grad_norm": 2.371277093887329,
|
| 1192 |
+
"learning_rate": 1.8778544091784047e-05,
|
| 1193 |
+
"loss": 0.2293,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.430769230769231,
|
| 1198 |
+
"eval_loss": 0.8893064260482788,
|
| 1199 |
+
"eval_runtime": 1.1684,
|
| 1200 |
+
"eval_samples_per_second": 97.567,
|
| 1201 |
+
"eval_steps_per_second": 5.135,
|
| 1202 |
+
"step": 790
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.4615384615384617,
|
| 1206 |
+
"grad_norm": 1.4892629384994507,
|
| 1207 |
+
"learning_rate": 1.8726599588756144e-05,
|
| 1208 |
+
"loss": 0.2326,
|
| 1209 |
+
"step": 800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.4615384615384617,
|
| 1213 |
+
"eval_loss": 0.8676194548606873,
|
| 1214 |
+
"eval_runtime": 1.1699,
|
| 1215 |
+
"eval_samples_per_second": 97.441,
|
| 1216 |
+
"eval_steps_per_second": 5.128,
|
| 1217 |
+
"step": 800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.4923076923076923,
|
| 1221 |
+
"grad_norm": 1.6047152280807495,
|
| 1222 |
+
"learning_rate": 1.8673648411665895e-05,
|
| 1223 |
+
"loss": 0.1986,
|
| 1224 |
+
"step": 810
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.4923076923076923,
|
| 1228 |
+
"eval_loss": 0.8942193984985352,
|
| 1229 |
+
"eval_runtime": 1.17,
|
| 1230 |
+
"eval_samples_per_second": 97.433,
|
| 1231 |
+
"eval_steps_per_second": 5.128,
|
| 1232 |
+
"step": 810
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.523076923076923,
|
| 1236 |
+
"grad_norm": 2.106461524963379,
|
| 1237 |
+
"learning_rate": 1.8619696668800494e-05,
|
| 1238 |
+
"loss": 0.238,
|
| 1239 |
+
"step": 820
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.523076923076923,
|
| 1243 |
+
"eval_loss": 0.8907297849655151,
|
| 1244 |
+
"eval_runtime": 1.1681,
|
| 1245 |
+
"eval_samples_per_second": 97.593,
|
| 1246 |
+
"eval_steps_per_second": 5.136,
|
| 1247 |
+
"step": 820
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 2.5538461538461537,
|
| 1251 |
+
"grad_norm": 1.7706941366195679,
|
| 1252 |
+
"learning_rate": 1.8564750583869374e-05,
|
| 1253 |
+
"loss": 0.2278,
|
| 1254 |
+
"step": 830
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 2.5538461538461537,
|
| 1258 |
+
"eval_loss": 0.8588199019432068,
|
| 1259 |
+
"eval_runtime": 1.1709,
|
| 1260 |
+
"eval_samples_per_second": 97.36,
|
| 1261 |
+
"eval_steps_per_second": 5.124,
|
| 1262 |
+
"step": 830
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.5846153846153848,
|
| 1266 |
+
"grad_norm": 1.4446183443069458,
|
| 1267 |
+
"learning_rate": 1.850881649528625e-05,
|
| 1268 |
+
"loss": 0.2033,
|
| 1269 |
+
"step": 840
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.5846153846153848,
|
| 1273 |
+
"eval_loss": 0.8720155954360962,
|
| 1274 |
+
"eval_runtime": 1.1716,
|
| 1275 |
+
"eval_samples_per_second": 97.304,
|
| 1276 |
+
"eval_steps_per_second": 5.121,
|
| 1277 |
+
"step": 840
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.6153846153846154,
|
| 1281 |
+
"grad_norm": 1.5155885219573975,
|
| 1282 |
+
"learning_rate": 1.845190085543795e-05,
|
| 1283 |
+
"loss": 0.2211,
|
| 1284 |
+
"step": 850
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6153846153846154,
|
| 1288 |
+
"eval_loss": 0.8732415437698364,
|
| 1289 |
+
"eval_runtime": 1.1708,
|
| 1290 |
+
"eval_samples_per_second": 97.371,
|
| 1291 |
+
"eval_steps_per_second": 5.125,
|
| 1292 |
+
"step": 850
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.646153846153846,
|
| 1296 |
+
"grad_norm": 2.934957981109619,
|
| 1297 |
+
"learning_rate": 1.839401022994006e-05,
|
| 1298 |
+
"loss": 0.21,
|
| 1299 |
+
"step": 860
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.646153846153846,
|
| 1303 |
+
"eval_loss": 0.8790932297706604,
|
| 1304 |
+
"eval_runtime": 1.1671,
|
| 1305 |
+
"eval_samples_per_second": 97.68,
|
| 1306 |
+
"eval_steps_per_second": 5.141,
|
| 1307 |
+
"step": 860
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.676923076923077,
|
| 1311 |
+
"grad_norm": 2.930457353591919,
|
| 1312 |
+
"learning_rate": 1.8335151296879576e-05,
|
| 1313 |
+
"loss": 0.2313,
|
| 1314 |
+
"step": 870
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.676923076923077,
|
| 1318 |
+
"eval_loss": 0.8555310368537903,
|
| 1319 |
+
"eval_runtime": 1.1718,
|
| 1320 |
+
"eval_samples_per_second": 97.284,
|
| 1321 |
+
"eval_steps_per_second": 5.12,
|
| 1322 |
+
"step": 870
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.707692307692308,
|
| 1326 |
+
"grad_norm": 2.029303789138794,
|
| 1327 |
+
"learning_rate": 1.82753308460445e-05,
|
| 1328 |
+
"loss": 0.2285,
|
| 1329 |
+
"step": 880
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.707692307692308,
|
| 1333 |
+
"eval_loss": 0.870873749256134,
|
| 1334 |
+
"eval_runtime": 1.1696,
|
| 1335 |
+
"eval_samples_per_second": 97.47,
|
| 1336 |
+
"eval_steps_per_second": 5.13,
|
| 1337 |
+
"step": 880
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 2.7384615384615385,
|
| 1341 |
+
"grad_norm": 1.949743390083313,
|
| 1342 |
+
"learning_rate": 1.821455577814062e-05,
|
| 1343 |
+
"loss": 0.2134,
|
| 1344 |
+
"step": 890
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.7384615384615385,
|
| 1348 |
+
"eval_loss": 0.8873838186264038,
|
| 1349 |
+
"eval_runtime": 1.1688,
|
| 1350 |
+
"eval_samples_per_second": 97.538,
|
| 1351 |
+
"eval_steps_per_second": 5.134,
|
| 1352 |
+
"step": 890
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.769230769230769,
|
| 1356 |
+
"grad_norm": 1.6945637464523315,
|
| 1357 |
+
"learning_rate": 1.8152833103995443e-05,
|
| 1358 |
+
"loss": 0.218,
|
| 1359 |
+
"step": 900
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.769230769230769,
|
| 1363 |
+
"eval_loss": 0.8798539042472839,
|
| 1364 |
+
"eval_runtime": 1.1699,
|
| 1365 |
+
"eval_samples_per_second": 97.444,
|
| 1366 |
+
"eval_steps_per_second": 5.129,
|
| 1367 |
+
"step": 900
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8,
|
| 1371 |
+
"grad_norm": 2.2784035205841064,
|
| 1372 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 1373 |
+
"loss": 0.2426,
|
| 1374 |
+
"step": 910
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8,
|
| 1378 |
+
"eval_loss": 0.8753871321678162,
|
| 1379 |
+
"eval_runtime": 1.1723,
|
| 1380 |
+
"eval_samples_per_second": 97.246,
|
| 1381 |
+
"eval_steps_per_second": 5.118,
|
| 1382 |
+
"step": 910
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.830769230769231,
|
| 1386 |
+
"grad_norm": 1.750718593597412,
|
| 1387 |
+
"learning_rate": 1.802657352603483e-05,
|
| 1388 |
+
"loss": 0.2059,
|
| 1389 |
+
"step": 920
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.830769230769231,
|
| 1393 |
+
"eval_loss": 0.8761821985244751,
|
| 1394 |
+
"eval_runtime": 1.1762,
|
| 1395 |
+
"eval_samples_per_second": 96.92,
|
| 1396 |
+
"eval_steps_per_second": 5.101,
|
| 1397 |
+
"step": 920
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.8615384615384616,
|
| 1401 |
+
"grad_norm": 1.2204371690750122,
|
| 1402 |
+
"learning_rate": 1.7962051187141377e-05,
|
| 1403 |
+
"loss": 0.2229,
|
| 1404 |
+
"step": 930
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.8615384615384616,
|
| 1408 |
+
"eval_loss": 0.8772630095481873,
|
| 1409 |
+
"eval_runtime": 1.1697,
|
| 1410 |
+
"eval_samples_per_second": 97.458,
|
| 1411 |
+
"eval_steps_per_second": 5.129,
|
| 1412 |
+
"step": 930
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.8923076923076922,
|
| 1416 |
+
"grad_norm": 2.0748281478881836,
|
| 1417 |
+
"learning_rate": 1.7896610370170452e-05,
|
| 1418 |
+
"loss": 0.2582,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.8923076923076922,
|
| 1423 |
+
"eval_loss": 0.884286105632782,
|
| 1424 |
+
"eval_runtime": 1.1773,
|
| 1425 |
+
"eval_samples_per_second": 96.829,
|
| 1426 |
+
"eval_steps_per_second": 5.096,
|
| 1427 |
+
"step": 940
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.9230769230769234,
|
| 1431 |
+
"grad_norm": 1.9051027297973633,
|
| 1432 |
+
"learning_rate": 1.7830258624176224e-05,
|
| 1433 |
+
"loss": 0.2102,
|
| 1434 |
+
"step": 950
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.9230769230769234,
|
| 1438 |
+
"eval_loss": 0.878767192363739,
|
| 1439 |
+
"eval_runtime": 1.1684,
|
| 1440 |
+
"eval_samples_per_second": 97.571,
|
| 1441 |
+
"eval_steps_per_second": 5.135,
|
| 1442 |
+
"step": 950
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.953846153846154,
|
| 1446 |
+
"grad_norm": 1.8921849727630615,
|
| 1447 |
+
"learning_rate": 1.776300360329488e-05,
|
| 1448 |
+
"loss": 0.2331,
|
| 1449 |
+
"step": 960
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.953846153846154,
|
| 1453 |
+
"eval_loss": 0.8681402206420898,
|
| 1454 |
+
"eval_runtime": 1.1693,
|
| 1455 |
+
"eval_samples_per_second": 97.497,
|
| 1456 |
+
"eval_steps_per_second": 5.131,
|
| 1457 |
+
"step": 960
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.9846153846153847,
|
| 1461 |
+
"grad_norm": 1.5593761205673218,
|
| 1462 |
+
"learning_rate": 1.769485306586166e-05,
|
| 1463 |
+
"loss": 0.2226,
|
| 1464 |
+
"step": 970
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.9846153846153847,
|
| 1468 |
+
"eval_loss": 0.8726215958595276,
|
| 1469 |
+
"eval_runtime": 1.1712,
|
| 1470 |
+
"eval_samples_per_second": 97.336,
|
| 1471 |
+
"eval_steps_per_second": 5.123,
|
| 1472 |
+
"step": 970
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.0153846153846153,
|
| 1476 |
+
"grad_norm": 1.4075913429260254,
|
| 1477 |
+
"learning_rate": 1.762581487351587e-05,
|
| 1478 |
+
"loss": 0.1535,
|
| 1479 |
+
"step": 980
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.0153846153846153,
|
| 1483 |
+
"eval_loss": 0.9113463163375854,
|
| 1484 |
+
"eval_runtime": 1.1794,
|
| 1485 |
+
"eval_samples_per_second": 96.661,
|
| 1486 |
+
"eval_steps_per_second": 5.087,
|
| 1487 |
+
"step": 980
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.046153846153846,
|
| 1491 |
+
"grad_norm": 1.410906434059143,
|
| 1492 |
+
"learning_rate": 1.7555896990294003e-05,
|
| 1493 |
+
"loss": 0.1021,
|
| 1494 |
+
"step": 990
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.046153846153846,
|
| 1498 |
+
"eval_loss": 0.9633604288101196,
|
| 1499 |
+
"eval_runtime": 1.1702,
|
| 1500 |
+
"eval_samples_per_second": 97.416,
|
| 1501 |
+
"eval_steps_per_second": 5.127,
|
| 1502 |
+
"step": 990
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 3.076923076923077,
|
| 1506 |
+
"grad_norm": 1.0470328330993652,
|
| 1507 |
+
"learning_rate": 1.7485107481711014e-05,
|
| 1508 |
+
"loss": 0.1118,
|
| 1509 |
+
"step": 1000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 3.076923076923077,
|
| 1513 |
+
"eval_loss": 0.9647616744041443,
|
| 1514 |
+
"eval_runtime": 1.1683,
|
| 1515 |
+
"eval_samples_per_second": 97.574,
|
| 1516 |
+
"eval_steps_per_second": 5.135,
|
| 1517 |
+
"step": 1000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.1076923076923078,
|
| 1521 |
+
"grad_norm": 2.098592758178711,
|
| 1522 |
+
"learning_rate": 1.741345451382992e-05,
|
| 1523 |
+
"loss": 0.1224,
|
| 1524 |
+
"step": 1010
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.1076923076923078,
|
| 1528 |
+
"eval_loss": 0.9478728175163269,
|
| 1529 |
+
"eval_runtime": 1.1735,
|
| 1530 |
+
"eval_samples_per_second": 97.143,
|
| 1531 |
+
"eval_steps_per_second": 5.113,
|
| 1532 |
+
"step": 1010
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 3.1384615384615384,
|
| 1536 |
+
"grad_norm": 1.6263700723648071,
|
| 1537 |
+
"learning_rate": 1.7340946352319795e-05,
|
| 1538 |
+
"loss": 0.1377,
|
| 1539 |
+
"step": 1020
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 3.1384615384615384,
|
| 1543 |
+
"eval_loss": 0.922594428062439,
|
| 1544 |
+
"eval_runtime": 1.1721,
|
| 1545 |
+
"eval_samples_per_second": 97.262,
|
| 1546 |
+
"eval_steps_per_second": 5.119,
|
| 1547 |
+
"step": 1020
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 3.169230769230769,
|
| 1551 |
+
"grad_norm": 1.0413360595703125,
|
| 1552 |
+
"learning_rate": 1.7267591361502233e-05,
|
| 1553 |
+
"loss": 0.1249,
|
| 1554 |
+
"step": 1030
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 3.169230769230769,
|
| 1558 |
+
"eval_loss": 0.9370183944702148,
|
| 1559 |
+
"eval_runtime": 1.169,
|
| 1560 |
+
"eval_samples_per_second": 97.519,
|
| 1561 |
+
"eval_steps_per_second": 5.133,
|
| 1562 |
+
"step": 1030
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 3.2,
|
| 1566 |
+
"grad_norm": 1.6745877265930176,
|
| 1567 |
+
"learning_rate": 1.7193398003386514e-05,
|
| 1568 |
+
"loss": 0.1186,
|
| 1569 |
+
"step": 1040
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 3.2,
|
| 1573 |
+
"eval_loss": 0.9573487639427185,
|
| 1574 |
+
"eval_runtime": 1.1681,
|
| 1575 |
+
"eval_samples_per_second": 97.596,
|
| 1576 |
+
"eval_steps_per_second": 5.137,
|
| 1577 |
+
"step": 1040
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.230769230769231,
|
| 1581 |
+
"grad_norm": 1.3979861736297607,
|
| 1582 |
+
"learning_rate": 1.7118374836693407e-05,
|
| 1583 |
+
"loss": 0.135,
|
| 1584 |
+
"step": 1050
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.230769230769231,
|
| 1588 |
+
"eval_loss": 0.9554935693740845,
|
| 1589 |
+
"eval_runtime": 1.1664,
|
| 1590 |
+
"eval_samples_per_second": 97.735,
|
| 1591 |
+
"eval_steps_per_second": 5.144,
|
| 1592 |
+
"step": 1050
|
| 1593 |
+
}
|
| 1594 |
+
],
|
| 1595 |
+
"logging_steps": 10,
|
| 1596 |
+
"max_steps": 3250,
|
| 1597 |
+
"num_input_tokens_seen": 0,
|
| 1598 |
+
"num_train_epochs": 10,
|
| 1599 |
+
"save_steps": 50,
|
| 1600 |
+
"total_flos": 6.558281165386547e+16,
|
| 1601 |
+
"train_batch_size": 2,
|
| 1602 |
+
"trial_name": null,
|
| 1603 |
+
"trial_params": null
|
| 1604 |
+
}
|
{checkpoint-150 → checkpoint-1050}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/added_tokens.json
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/config.json
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/generation_config.json
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/model.safetensors.index.json
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/special_tokens_map.json
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/tokenizer.json
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/tokenizer.model
RENAMED
|
File without changes
|
{checkpoint-200 → checkpoint-1100}/tokenizer_config.json
RENAMED
|
File without changes
|
checkpoint-1100/trainer_state.json
ADDED
|
@@ -0,0 +1,1679 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.3846153846153846,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006153846153846154,
|
| 13 |
+
"eval_loss": 1.487821340560913,
|
| 14 |
+
"eval_runtime": 1.2374,
|
| 15 |
+
"eval_samples_per_second": 92.131,
|
| 16 |
+
"eval_steps_per_second": 4.849,
|
| 17 |
+
"step": 2
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03076923076923077,
|
| 21 |
+
"grad_norm": 20.007999420166016,
|
| 22 |
+
"learning_rate": 6.153846153846155e-07,
|
| 23 |
+
"loss": 1.4703,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03076923076923077,
|
| 28 |
+
"eval_loss": 1.4221965074539185,
|
| 29 |
+
"eval_runtime": 1.1701,
|
| 30 |
+
"eval_samples_per_second": 97.431,
|
| 31 |
+
"eval_steps_per_second": 5.128,
|
| 32 |
+
"step": 10
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06153846153846154,
|
| 36 |
+
"grad_norm": 29.657033920288086,
|
| 37 |
+
"learning_rate": 1.230769230769231e-06,
|
| 38 |
+
"loss": 1.4402,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06153846153846154,
|
| 43 |
+
"eval_loss": 1.238189935684204,
|
| 44 |
+
"eval_runtime": 1.1729,
|
| 45 |
+
"eval_samples_per_second": 97.195,
|
| 46 |
+
"eval_steps_per_second": 5.116,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.09230769230769231,
|
| 51 |
+
"grad_norm": 10.80688190460205,
|
| 52 |
+
"learning_rate": 1.8461538461538465e-06,
|
| 53 |
+
"loss": 1.0957,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09230769230769231,
|
| 58 |
+
"eval_loss": 0.908875048160553,
|
| 59 |
+
"eval_runtime": 1.1739,
|
| 60 |
+
"eval_samples_per_second": 97.111,
|
| 61 |
+
"eval_steps_per_second": 5.111,
|
| 62 |
+
"step": 30
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.12307692307692308,
|
| 66 |
+
"grad_norm": 12.914667129516602,
|
| 67 |
+
"learning_rate": 2.461538461538462e-06,
|
| 68 |
+
"loss": 0.8136,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.12307692307692308,
|
| 73 |
+
"eval_loss": 0.7436326146125793,
|
| 74 |
+
"eval_runtime": 1.1712,
|
| 75 |
+
"eval_samples_per_second": 97.34,
|
| 76 |
+
"eval_steps_per_second": 5.123,
|
| 77 |
+
"step": 40
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.15384615384615385,
|
| 81 |
+
"grad_norm": 5.206438064575195,
|
| 82 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 83 |
+
"loss": 0.7864,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.15384615384615385,
|
| 88 |
+
"eval_loss": 0.6924600601196289,
|
| 89 |
+
"eval_runtime": 1.1718,
|
| 90 |
+
"eval_samples_per_second": 97.287,
|
| 91 |
+
"eval_steps_per_second": 5.12,
|
| 92 |
+
"step": 50
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.18461538461538463,
|
| 96 |
+
"grad_norm": 5.670599460601807,
|
| 97 |
+
"learning_rate": 3.692307692307693e-06,
|
| 98 |
+
"loss": 0.622,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.18461538461538463,
|
| 103 |
+
"eval_loss": 0.6791077852249146,
|
| 104 |
+
"eval_runtime": 1.1678,
|
| 105 |
+
"eval_samples_per_second": 97.62,
|
| 106 |
+
"eval_steps_per_second": 5.138,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.2153846153846154,
|
| 111 |
+
"grad_norm": 3.6266868114471436,
|
| 112 |
+
"learning_rate": 4.307692307692308e-06,
|
| 113 |
+
"loss": 0.6672,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2153846153846154,
|
| 118 |
+
"eval_loss": 0.6631948947906494,
|
| 119 |
+
"eval_runtime": 1.1626,
|
| 120 |
+
"eval_samples_per_second": 98.06,
|
| 121 |
+
"eval_steps_per_second": 5.161,
|
| 122 |
+
"step": 70
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.24615384615384617,
|
| 126 |
+
"grad_norm": 3.1027884483337402,
|
| 127 |
+
"learning_rate": 4.923076923076924e-06,
|
| 128 |
+
"loss": 0.6097,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.24615384615384617,
|
| 133 |
+
"eval_loss": 0.6625475883483887,
|
| 134 |
+
"eval_runtime": 1.1717,
|
| 135 |
+
"eval_samples_per_second": 97.294,
|
| 136 |
+
"eval_steps_per_second": 5.121,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.27692307692307694,
|
| 141 |
+
"grad_norm": 3.233839511871338,
|
| 142 |
+
"learning_rate": 5.538461538461539e-06,
|
| 143 |
+
"loss": 0.6957,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.27692307692307694,
|
| 148 |
+
"eval_loss": 0.6590337157249451,
|
| 149 |
+
"eval_runtime": 1.1664,
|
| 150 |
+
"eval_samples_per_second": 97.738,
|
| 151 |
+
"eval_steps_per_second": 5.144,
|
| 152 |
+
"step": 90
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.3076923076923077,
|
| 156 |
+
"grad_norm": 4.932614803314209,
|
| 157 |
+
"learning_rate": 6.153846153846155e-06,
|
| 158 |
+
"loss": 0.6994,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.3076923076923077,
|
| 163 |
+
"eval_loss": 0.6592673659324646,
|
| 164 |
+
"eval_runtime": 1.1692,
|
| 165 |
+
"eval_samples_per_second": 97.502,
|
| 166 |
+
"eval_steps_per_second": 5.132,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.3384615384615385,
|
| 171 |
+
"grad_norm": 2.642303466796875,
|
| 172 |
+
"learning_rate": 6.76923076923077e-06,
|
| 173 |
+
"loss": 0.6501,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.3384615384615385,
|
| 178 |
+
"eval_loss": 0.6589992642402649,
|
| 179 |
+
"eval_runtime": 1.1669,
|
| 180 |
+
"eval_samples_per_second": 97.691,
|
| 181 |
+
"eval_steps_per_second": 5.142,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.36923076923076925,
|
| 186 |
+
"grad_norm": 3.45409893989563,
|
| 187 |
+
"learning_rate": 7.384615384615386e-06,
|
| 188 |
+
"loss": 0.7512,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.36923076923076925,
|
| 193 |
+
"eval_loss": 0.6568591594696045,
|
| 194 |
+
"eval_runtime": 1.1778,
|
| 195 |
+
"eval_samples_per_second": 96.79,
|
| 196 |
+
"eval_steps_per_second": 5.094,
|
| 197 |
+
"step": 120
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.4,
|
| 201 |
+
"grad_norm": 8.460351943969727,
|
| 202 |
+
"learning_rate": 8.000000000000001e-06,
|
| 203 |
+
"loss": 0.5996,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4,
|
| 208 |
+
"eval_loss": 0.6578297019004822,
|
| 209 |
+
"eval_runtime": 1.1683,
|
| 210 |
+
"eval_samples_per_second": 97.575,
|
| 211 |
+
"eval_steps_per_second": 5.136,
|
| 212 |
+
"step": 130
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4307692307692308,
|
| 216 |
+
"grad_norm": 2.4495465755462646,
|
| 217 |
+
"learning_rate": 8.615384615384617e-06,
|
| 218 |
+
"loss": 0.6498,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4307692307692308,
|
| 223 |
+
"eval_loss": 0.6588199138641357,
|
| 224 |
+
"eval_runtime": 1.1664,
|
| 225 |
+
"eval_samples_per_second": 97.739,
|
| 226 |
+
"eval_steps_per_second": 5.144,
|
| 227 |
+
"step": 140
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.46153846153846156,
|
| 231 |
+
"grad_norm": 2.6779162883758545,
|
| 232 |
+
"learning_rate": 9.230769230769232e-06,
|
| 233 |
+
"loss": 0.703,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.46153846153846156,
|
| 238 |
+
"eval_loss": 0.6626676321029663,
|
| 239 |
+
"eval_runtime": 1.1764,
|
| 240 |
+
"eval_samples_per_second": 96.904,
|
| 241 |
+
"eval_steps_per_second": 5.1,
|
| 242 |
+
"step": 150
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.49230769230769234,
|
| 246 |
+
"grad_norm": 8.147880554199219,
|
| 247 |
+
"learning_rate": 9.846153846153848e-06,
|
| 248 |
+
"loss": 0.6649,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.49230769230769234,
|
| 253 |
+
"eval_loss": 0.6644802689552307,
|
| 254 |
+
"eval_runtime": 1.1701,
|
| 255 |
+
"eval_samples_per_second": 97.424,
|
| 256 |
+
"eval_steps_per_second": 5.128,
|
| 257 |
+
"step": 160
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.5230769230769231,
|
| 261 |
+
"grad_norm": 2.7410728931427,
|
| 262 |
+
"learning_rate": 1.0461538461538463e-05,
|
| 263 |
+
"loss": 0.6704,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.5230769230769231,
|
| 268 |
+
"eval_loss": 0.6591455340385437,
|
| 269 |
+
"eval_runtime": 1.1722,
|
| 270 |
+
"eval_samples_per_second": 97.253,
|
| 271 |
+
"eval_steps_per_second": 5.119,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.5538461538461539,
|
| 276 |
+
"grad_norm": 6.886598587036133,
|
| 277 |
+
"learning_rate": 1.1076923076923079e-05,
|
| 278 |
+
"loss": 0.6073,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.5538461538461539,
|
| 283 |
+
"eval_loss": 0.6617754697799683,
|
| 284 |
+
"eval_runtime": 1.1682,
|
| 285 |
+
"eval_samples_per_second": 97.588,
|
| 286 |
+
"eval_steps_per_second": 5.136,
|
| 287 |
+
"step": 180
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.5846153846153846,
|
| 291 |
+
"grad_norm": 4.26249361038208,
|
| 292 |
+
"learning_rate": 1.1692307692307694e-05,
|
| 293 |
+
"loss": 0.7556,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.5846153846153846,
|
| 298 |
+
"eval_loss": 0.6659030318260193,
|
| 299 |
+
"eval_runtime": 1.1721,
|
| 300 |
+
"eval_samples_per_second": 97.265,
|
| 301 |
+
"eval_steps_per_second": 5.119,
|
| 302 |
+
"step": 190
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.6153846153846154,
|
| 306 |
+
"grad_norm": 4.204222202301025,
|
| 307 |
+
"learning_rate": 1.230769230769231e-05,
|
| 308 |
+
"loss": 0.6583,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.6153846153846154,
|
| 313 |
+
"eval_loss": 0.6660778522491455,
|
| 314 |
+
"eval_runtime": 1.1728,
|
| 315 |
+
"eval_samples_per_second": 97.201,
|
| 316 |
+
"eval_steps_per_second": 5.116,
|
| 317 |
+
"step": 200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6461538461538462,
|
| 321 |
+
"grad_norm": 3.154658079147339,
|
| 322 |
+
"learning_rate": 1.2923076923076925e-05,
|
| 323 |
+
"loss": 0.7288,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6461538461538462,
|
| 328 |
+
"eval_loss": 0.6642422676086426,
|
| 329 |
+
"eval_runtime": 1.166,
|
| 330 |
+
"eval_samples_per_second": 97.772,
|
| 331 |
+
"eval_steps_per_second": 5.146,
|
| 332 |
+
"step": 210
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.676923076923077,
|
| 336 |
+
"grad_norm": 3.248359203338623,
|
| 337 |
+
"learning_rate": 1.353846153846154e-05,
|
| 338 |
+
"loss": 0.6702,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.676923076923077,
|
| 343 |
+
"eval_loss": 0.6680281758308411,
|
| 344 |
+
"eval_runtime": 1.1691,
|
| 345 |
+
"eval_samples_per_second": 97.507,
|
| 346 |
+
"eval_steps_per_second": 5.132,
|
| 347 |
+
"step": 220
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.7076923076923077,
|
| 351 |
+
"grad_norm": 2.5507216453552246,
|
| 352 |
+
"learning_rate": 1.4153846153846156e-05,
|
| 353 |
+
"loss": 0.66,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.7076923076923077,
|
| 358 |
+
"eval_loss": 0.6713310480117798,
|
| 359 |
+
"eval_runtime": 1.167,
|
| 360 |
+
"eval_samples_per_second": 97.683,
|
| 361 |
+
"eval_steps_per_second": 5.141,
|
| 362 |
+
"step": 230
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.7384615384615385,
|
| 366 |
+
"grad_norm": 2.1180007457733154,
|
| 367 |
+
"learning_rate": 1.4769230769230772e-05,
|
| 368 |
+
"loss": 0.6172,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.7384615384615385,
|
| 373 |
+
"eval_loss": 0.6759226322174072,
|
| 374 |
+
"eval_runtime": 1.1758,
|
| 375 |
+
"eval_samples_per_second": 96.954,
|
| 376 |
+
"eval_steps_per_second": 5.103,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.7692307692307693,
|
| 381 |
+
"grad_norm": 2.8318371772766113,
|
| 382 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 383 |
+
"loss": 0.5954,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.7692307692307693,
|
| 388 |
+
"eval_loss": 0.6854314208030701,
|
| 389 |
+
"eval_runtime": 1.1698,
|
| 390 |
+
"eval_samples_per_second": 97.457,
|
| 391 |
+
"eval_steps_per_second": 5.129,
|
| 392 |
+
"step": 250
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.8,
|
| 396 |
+
"grad_norm": 3.519829034805298,
|
| 397 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 398 |
+
"loss": 0.6639,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.8,
|
| 403 |
+
"eval_loss": 0.6849851012229919,
|
| 404 |
+
"eval_runtime": 1.1732,
|
| 405 |
+
"eval_samples_per_second": 97.173,
|
| 406 |
+
"eval_steps_per_second": 5.114,
|
| 407 |
+
"step": 260
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.8307692307692308,
|
| 411 |
+
"grad_norm": 3.7909388542175293,
|
| 412 |
+
"learning_rate": 1.6615384615384618e-05,
|
| 413 |
+
"loss": 0.6892,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.8307692307692308,
|
| 418 |
+
"eval_loss": 0.6863855123519897,
|
| 419 |
+
"eval_runtime": 1.1689,
|
| 420 |
+
"eval_samples_per_second": 97.53,
|
| 421 |
+
"eval_steps_per_second": 5.133,
|
| 422 |
+
"step": 270
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8615384615384616,
|
| 426 |
+
"grad_norm": 8.278326988220215,
|
| 427 |
+
"learning_rate": 1.7230769230769234e-05,
|
| 428 |
+
"loss": 0.6014,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8615384615384616,
|
| 433 |
+
"eval_loss": 0.6890373826026917,
|
| 434 |
+
"eval_runtime": 1.1722,
|
| 435 |
+
"eval_samples_per_second": 97.251,
|
| 436 |
+
"eval_steps_per_second": 5.118,
|
| 437 |
+
"step": 280
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8923076923076924,
|
| 441 |
+
"grad_norm": 2.616001605987549,
|
| 442 |
+
"learning_rate": 1.784615384615385e-05,
|
| 443 |
+
"loss": 0.6775,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.8923076923076924,
|
| 448 |
+
"eval_loss": 0.6797110438346863,
|
| 449 |
+
"eval_runtime": 1.1888,
|
| 450 |
+
"eval_samples_per_second": 95.897,
|
| 451 |
+
"eval_steps_per_second": 5.047,
|
| 452 |
+
"step": 290
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.9230769230769231,
|
| 456 |
+
"grad_norm": 2.4450528621673584,
|
| 457 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 458 |
+
"loss": 0.7434,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.9230769230769231,
|
| 463 |
+
"eval_loss": 0.6779448986053467,
|
| 464 |
+
"eval_runtime": 1.1722,
|
| 465 |
+
"eval_samples_per_second": 97.254,
|
| 466 |
+
"eval_steps_per_second": 5.119,
|
| 467 |
+
"step": 300
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.9538461538461539,
|
| 471 |
+
"grad_norm": 4.121164798736572,
|
| 472 |
+
"learning_rate": 1.907692307692308e-05,
|
| 473 |
+
"loss": 0.6413,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.9538461538461539,
|
| 478 |
+
"eval_loss": 0.6819297671318054,
|
| 479 |
+
"eval_runtime": 1.1683,
|
| 480 |
+
"eval_samples_per_second": 97.574,
|
| 481 |
+
"eval_steps_per_second": 5.135,
|
| 482 |
+
"step": 310
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.9846153846153847,
|
| 486 |
+
"grad_norm": 2.057229995727539,
|
| 487 |
+
"learning_rate": 1.9692307692307696e-05,
|
| 488 |
+
"loss": 0.6511,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.9846153846153847,
|
| 493 |
+
"eval_loss": 0.6861481666564941,
|
| 494 |
+
"eval_runtime": 1.1699,
|
| 495 |
+
"eval_samples_per_second": 97.444,
|
| 496 |
+
"eval_steps_per_second": 5.129,
|
| 497 |
+
"step": 320
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.0153846153846153,
|
| 501 |
+
"grad_norm": 1.7295210361480713,
|
| 502 |
+
"learning_rate": 1.9999855802751384e-05,
|
| 503 |
+
"loss": 0.5872,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.0153846153846153,
|
| 508 |
+
"eval_loss": 0.7481781244277954,
|
| 509 |
+
"eval_runtime": 1.172,
|
| 510 |
+
"eval_samples_per_second": 97.273,
|
| 511 |
+
"eval_steps_per_second": 5.12,
|
| 512 |
+
"step": 330
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.0461538461538462,
|
| 516 |
+
"grad_norm": 3.6910336017608643,
|
| 517 |
+
"learning_rate": 1.9998702249713747e-05,
|
| 518 |
+
"loss": 0.438,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.0461538461538462,
|
| 523 |
+
"eval_loss": 0.7471169829368591,
|
| 524 |
+
"eval_runtime": 1.175,
|
| 525 |
+
"eval_samples_per_second": 97.019,
|
| 526 |
+
"eval_steps_per_second": 5.106,
|
| 527 |
+
"step": 340
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.0769230769230769,
|
| 531 |
+
"grad_norm": 3.9220831394195557,
|
| 532 |
+
"learning_rate": 1.9996395276708856e-05,
|
| 533 |
+
"loss": 0.4141,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0769230769230769,
|
| 538 |
+
"eval_loss": 0.7425897121429443,
|
| 539 |
+
"eval_runtime": 1.1888,
|
| 540 |
+
"eval_samples_per_second": 95.893,
|
| 541 |
+
"eval_steps_per_second": 5.047,
|
| 542 |
+
"step": 350
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.1076923076923078,
|
| 546 |
+
"grad_norm": 2.1136879920959473,
|
| 547 |
+
"learning_rate": 1.9992935149862116e-05,
|
| 548 |
+
"loss": 0.4173,
|
| 549 |
+
"step": 360
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1076923076923078,
|
| 553 |
+
"eval_loss": 0.7477542161941528,
|
| 554 |
+
"eval_runtime": 1.1705,
|
| 555 |
+
"eval_samples_per_second": 97.398,
|
| 556 |
+
"eval_steps_per_second": 5.126,
|
| 557 |
+
"step": 360
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.1384615384615384,
|
| 561 |
+
"grad_norm": 2.187299966812134,
|
| 562 |
+
"learning_rate": 1.998832226832327e-05,
|
| 563 |
+
"loss": 0.4242,
|
| 564 |
+
"step": 370
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.1384615384615384,
|
| 568 |
+
"eval_loss": 0.7475994229316711,
|
| 569 |
+
"eval_runtime": 1.1683,
|
| 570 |
+
"eval_samples_per_second": 97.582,
|
| 571 |
+
"eval_steps_per_second": 5.136,
|
| 572 |
+
"step": 370
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.1692307692307693,
|
| 576 |
+
"grad_norm": 2.683750629425049,
|
| 577 |
+
"learning_rate": 1.9982557164220335e-05,
|
| 578 |
+
"loss": 0.4413,
|
| 579 |
+
"step": 380
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.1692307692307693,
|
| 583 |
+
"eval_loss": 0.7552060484886169,
|
| 584 |
+
"eval_runtime": 1.1689,
|
| 585 |
+
"eval_samples_per_second": 97.529,
|
| 586 |
+
"eval_steps_per_second": 5.133,
|
| 587 |
+
"step": 380
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.2,
|
| 591 |
+
"grad_norm": 3.312657117843628,
|
| 592 |
+
"learning_rate": 1.9975640502598243e-05,
|
| 593 |
+
"loss": 0.3361,
|
| 594 |
+
"step": 390
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.2,
|
| 598 |
+
"eval_loss": 0.7630029916763306,
|
| 599 |
+
"eval_runtime": 1.1689,
|
| 600 |
+
"eval_samples_per_second": 97.529,
|
| 601 |
+
"eval_steps_per_second": 5.133,
|
| 602 |
+
"step": 390
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.2307692307692308,
|
| 606 |
+
"grad_norm": 2.3290536403656006,
|
| 607 |
+
"learning_rate": 1.9967573081342103e-05,
|
| 608 |
+
"loss": 0.4174,
|
| 609 |
+
"step": 400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.2307692307692308,
|
| 613 |
+
"eval_loss": 0.7669293880462646,
|
| 614 |
+
"eval_runtime": 1.1711,
|
| 615 |
+
"eval_samples_per_second": 97.345,
|
| 616 |
+
"eval_steps_per_second": 5.123,
|
| 617 |
+
"step": 400
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.2615384615384615,
|
| 621 |
+
"grad_norm": 2.4483697414398193,
|
| 622 |
+
"learning_rate": 1.9958355831085155e-05,
|
| 623 |
+
"loss": 0.4045,
|
| 624 |
+
"step": 410
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.2615384615384615,
|
| 628 |
+
"eval_loss": 0.7652024626731873,
|
| 629 |
+
"eval_runtime": 1.1697,
|
| 630 |
+
"eval_samples_per_second": 97.461,
|
| 631 |
+
"eval_steps_per_second": 5.13,
|
| 632 |
+
"step": 410
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.2923076923076924,
|
| 636 |
+
"grad_norm": 1.703616976737976,
|
| 637 |
+
"learning_rate": 1.9947989815101444e-05,
|
| 638 |
+
"loss": 0.4292,
|
| 639 |
+
"step": 420
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.2923076923076924,
|
| 643 |
+
"eval_loss": 0.7553891539573669,
|
| 644 |
+
"eval_runtime": 1.1719,
|
| 645 |
+
"eval_samples_per_second": 97.277,
|
| 646 |
+
"eval_steps_per_second": 5.12,
|
| 647 |
+
"step": 420
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.323076923076923,
|
| 651 |
+
"grad_norm": 1.6245399713516235,
|
| 652 |
+
"learning_rate": 1.9936476229183133e-05,
|
| 653 |
+
"loss": 0.4164,
|
| 654 |
+
"step": 430
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.323076923076923,
|
| 658 |
+
"eval_loss": 0.7711736559867859,
|
| 659 |
+
"eval_runtime": 1.1776,
|
| 660 |
+
"eval_samples_per_second": 96.806,
|
| 661 |
+
"eval_steps_per_second": 5.095,
|
| 662 |
+
"step": 430
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.353846153846154,
|
| 666 |
+
"grad_norm": 1.9511611461639404,
|
| 667 |
+
"learning_rate": 1.992381640150257e-05,
|
| 668 |
+
"loss": 0.4337,
|
| 669 |
+
"step": 440
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.353846153846154,
|
| 673 |
+
"eval_loss": 0.7438372373580933,
|
| 674 |
+
"eval_runtime": 1.1692,
|
| 675 |
+
"eval_samples_per_second": 97.5,
|
| 676 |
+
"eval_steps_per_second": 5.132,
|
| 677 |
+
"step": 440
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.3846153846153846,
|
| 681 |
+
"grad_norm": 2.466737747192383,
|
| 682 |
+
"learning_rate": 1.9910011792459086e-05,
|
| 683 |
+
"loss": 0.3712,
|
| 684 |
+
"step": 450
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.3846153846153846,
|
| 688 |
+
"eval_loss": 0.8045614361763,
|
| 689 |
+
"eval_runtime": 1.1689,
|
| 690 |
+
"eval_samples_per_second": 97.53,
|
| 691 |
+
"eval_steps_per_second": 5.133,
|
| 692 |
+
"step": 450
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.4153846153846155,
|
| 696 |
+
"grad_norm": 2.473851203918457,
|
| 697 |
+
"learning_rate": 1.9895063994510512e-05,
|
| 698 |
+
"loss": 0.516,
|
| 699 |
+
"step": 460
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.4153846153846155,
|
| 703 |
+
"eval_loss": 0.7463352084159851,
|
| 704 |
+
"eval_runtime": 1.1697,
|
| 705 |
+
"eval_samples_per_second": 97.464,
|
| 706 |
+
"eval_steps_per_second": 5.13,
|
| 707 |
+
"step": 460
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.4461538461538461,
|
| 711 |
+
"grad_norm": 2.757955312728882,
|
| 712 |
+
"learning_rate": 1.9878974731989487e-05,
|
| 713 |
+
"loss": 0.4965,
|
| 714 |
+
"step": 470
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.4461538461538461,
|
| 718 |
+
"eval_loss": 0.7571713924407959,
|
| 719 |
+
"eval_runtime": 1.1688,
|
| 720 |
+
"eval_samples_per_second": 97.534,
|
| 721 |
+
"eval_steps_per_second": 5.133,
|
| 722 |
+
"step": 470
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.476923076923077,
|
| 726 |
+
"grad_norm": 3.9884915351867676,
|
| 727 |
+
"learning_rate": 1.9861745860904538e-05,
|
| 728 |
+
"loss": 0.3973,
|
| 729 |
+
"step": 480
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.476923076923077,
|
| 733 |
+
"eval_loss": 0.7763081789016724,
|
| 734 |
+
"eval_runtime": 1.168,
|
| 735 |
+
"eval_samples_per_second": 97.605,
|
| 736 |
+
"eval_steps_per_second": 5.137,
|
| 737 |
+
"step": 480
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5076923076923077,
|
| 741 |
+
"grad_norm": 1.9474631547927856,
|
| 742 |
+
"learning_rate": 1.9843379368725978e-05,
|
| 743 |
+
"loss": 0.4625,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5076923076923077,
|
| 748 |
+
"eval_loss": 0.7595367431640625,
|
| 749 |
+
"eval_runtime": 1.1708,
|
| 750 |
+
"eval_samples_per_second": 97.37,
|
| 751 |
+
"eval_steps_per_second": 5.125,
|
| 752 |
+
"step": 490
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5384615384615383,
|
| 756 |
+
"grad_norm": 2.2907869815826416,
|
| 757 |
+
"learning_rate": 1.9823877374156647e-05,
|
| 758 |
+
"loss": 0.4371,
|
| 759 |
+
"step": 500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5384615384615383,
|
| 763 |
+
"eval_loss": 0.762865424156189,
|
| 764 |
+
"eval_runtime": 1.1766,
|
| 765 |
+
"eval_samples_per_second": 96.891,
|
| 766 |
+
"eval_steps_per_second": 5.1,
|
| 767 |
+
"step": 500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.5692307692307692,
|
| 771 |
+
"grad_norm": 3.067608594894409,
|
| 772 |
+
"learning_rate": 1.9803242126887496e-05,
|
| 773 |
+
"loss": 0.45,
|
| 774 |
+
"step": 510
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.5692307692307692,
|
| 778 |
+
"eval_loss": 0.7638718485832214,
|
| 779 |
+
"eval_runtime": 1.1684,
|
| 780 |
+
"eval_samples_per_second": 97.568,
|
| 781 |
+
"eval_steps_per_second": 5.135,
|
| 782 |
+
"step": 510
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.6,
|
| 786 |
+
"grad_norm": 2.1113197803497314,
|
| 787 |
+
"learning_rate": 1.9781476007338058e-05,
|
| 788 |
+
"loss": 0.4524,
|
| 789 |
+
"step": 520
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.6,
|
| 793 |
+
"eval_loss": 0.7560412883758545,
|
| 794 |
+
"eval_runtime": 1.1719,
|
| 795 |
+
"eval_samples_per_second": 97.278,
|
| 796 |
+
"eval_steps_per_second": 5.12,
|
| 797 |
+
"step": 520
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.6307692307692307,
|
| 801 |
+
"grad_norm": 1.889585256576538,
|
| 802 |
+
"learning_rate": 1.9758581526381878e-05,
|
| 803 |
+
"loss": 0.4287,
|
| 804 |
+
"step": 530
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.6307692307692307,
|
| 808 |
+
"eval_loss": 0.7591350674629211,
|
| 809 |
+
"eval_runtime": 1.1693,
|
| 810 |
+
"eval_samples_per_second": 97.495,
|
| 811 |
+
"eval_steps_per_second": 5.131,
|
| 812 |
+
"step": 530
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.6615384615384614,
|
| 816 |
+
"grad_norm": 2.519418716430664,
|
| 817 |
+
"learning_rate": 1.973456132505684e-05,
|
| 818 |
+
"loss": 0.4603,
|
| 819 |
+
"step": 540
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.6615384615384614,
|
| 823 |
+
"eval_loss": 0.7536805868148804,
|
| 824 |
+
"eval_runtime": 1.1799,
|
| 825 |
+
"eval_samples_per_second": 96.617,
|
| 826 |
+
"eval_steps_per_second": 5.085,
|
| 827 |
+
"step": 540
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.6923076923076923,
|
| 831 |
+
"grad_norm": 2.7455596923828125,
|
| 832 |
+
"learning_rate": 1.9709418174260523e-05,
|
| 833 |
+
"loss": 0.4293,
|
| 834 |
+
"step": 550
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.6923076923076923,
|
| 838 |
+
"eval_loss": 0.763169527053833,
|
| 839 |
+
"eval_runtime": 1.1696,
|
| 840 |
+
"eval_samples_per_second": 97.47,
|
| 841 |
+
"eval_steps_per_second": 5.13,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7230769230769232,
|
| 846 |
+
"grad_norm": 1.9823411703109741,
|
| 847 |
+
"learning_rate": 1.9683154974430544e-05,
|
| 848 |
+
"loss": 0.4643,
|
| 849 |
+
"step": 560
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7230769230769232,
|
| 853 |
+
"eval_loss": 0.7607314586639404,
|
| 854 |
+
"eval_runtime": 1.1712,
|
| 855 |
+
"eval_samples_per_second": 97.336,
|
| 856 |
+
"eval_steps_per_second": 5.123,
|
| 857 |
+
"step": 560
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.7538461538461538,
|
| 861 |
+
"grad_norm": 2.2437684535980225,
|
| 862 |
+
"learning_rate": 1.965577475520999e-05,
|
| 863 |
+
"loss": 0.452,
|
| 864 |
+
"step": 570
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.7538461538461538,
|
| 868 |
+
"eval_loss": 0.7615717649459839,
|
| 869 |
+
"eval_runtime": 1.1946,
|
| 870 |
+
"eval_samples_per_second": 95.425,
|
| 871 |
+
"eval_steps_per_second": 5.022,
|
| 872 |
+
"step": 570
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.7846153846153845,
|
| 876 |
+
"grad_norm": 3.0740182399749756,
|
| 877 |
+
"learning_rate": 1.962728067509791e-05,
|
| 878 |
+
"loss": 0.4885,
|
| 879 |
+
"step": 580
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.7846153846153845,
|
| 883 |
+
"eval_loss": 0.7764689326286316,
|
| 884 |
+
"eval_runtime": 1.1714,
|
| 885 |
+
"eval_samples_per_second": 97.32,
|
| 886 |
+
"eval_steps_per_second": 5.122,
|
| 887 |
+
"step": 580
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.8153846153846154,
|
| 891 |
+
"grad_norm": 2.050313711166382,
|
| 892 |
+
"learning_rate": 1.9597676021084962e-05,
|
| 893 |
+
"loss": 0.4992,
|
| 894 |
+
"step": 590
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.8153846153846154,
|
| 898 |
+
"eval_loss": 0.7457339763641357,
|
| 899 |
+
"eval_runtime": 1.1747,
|
| 900 |
+
"eval_samples_per_second": 97.045,
|
| 901 |
+
"eval_steps_per_second": 5.108,
|
| 902 |
+
"step": 590
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.8461538461538463,
|
| 906 |
+
"grad_norm": 2.349470853805542,
|
| 907 |
+
"learning_rate": 1.9566964208274254e-05,
|
| 908 |
+
"loss": 0.3963,
|
| 909 |
+
"step": 600
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.8461538461538463,
|
| 913 |
+
"eval_loss": 0.7712545990943909,
|
| 914 |
+
"eval_runtime": 1.17,
|
| 915 |
+
"eval_samples_per_second": 97.433,
|
| 916 |
+
"eval_steps_per_second": 5.128,
|
| 917 |
+
"step": 600
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.876923076923077,
|
| 921 |
+
"grad_norm": 1.947912573814392,
|
| 922 |
+
"learning_rate": 1.9535148779487365e-05,
|
| 923 |
+
"loss": 0.4388,
|
| 924 |
+
"step": 610
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.876923076923077,
|
| 928 |
+
"eval_loss": 0.772519052028656,
|
| 929 |
+
"eval_runtime": 1.1717,
|
| 930 |
+
"eval_samples_per_second": 97.296,
|
| 931 |
+
"eval_steps_per_second": 5.121,
|
| 932 |
+
"step": 610
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.9076923076923076,
|
| 936 |
+
"grad_norm": 1.8837225437164307,
|
| 937 |
+
"learning_rate": 1.9502233404855672e-05,
|
| 938 |
+
"loss": 0.3965,
|
| 939 |
+
"step": 620
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.9076923076923076,
|
| 943 |
+
"eval_loss": 0.7594517469406128,
|
| 944 |
+
"eval_runtime": 1.2215,
|
| 945 |
+
"eval_samples_per_second": 93.326,
|
| 946 |
+
"eval_steps_per_second": 4.912,
|
| 947 |
+
"step": 620
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9384615384615385,
|
| 951 |
+
"grad_norm": 1.8398792743682861,
|
| 952 |
+
"learning_rate": 1.946822188139696e-05,
|
| 953 |
+
"loss": 0.3898,
|
| 954 |
+
"step": 630
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.9384615384615385,
|
| 958 |
+
"eval_loss": 0.7748451828956604,
|
| 959 |
+
"eval_runtime": 1.2222,
|
| 960 |
+
"eval_samples_per_second": 93.273,
|
| 961 |
+
"eval_steps_per_second": 4.909,
|
| 962 |
+
"step": 630
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.9692307692307693,
|
| 966 |
+
"grad_norm": 2.3971219062805176,
|
| 967 |
+
"learning_rate": 1.9433118132577432e-05,
|
| 968 |
+
"loss": 0.4253,
|
| 969 |
+
"step": 640
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.9692307692307693,
|
| 973 |
+
"eval_loss": 0.7729252576828003,
|
| 974 |
+
"eval_runtime": 1.1743,
|
| 975 |
+
"eval_samples_per_second": 97.081,
|
| 976 |
+
"eval_steps_per_second": 5.11,
|
| 977 |
+
"step": 640
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.0,
|
| 981 |
+
"grad_norm": 1.657328724861145,
|
| 982 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 983 |
+
"loss": 0.4898,
|
| 984 |
+
"step": 650
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.0,
|
| 988 |
+
"eval_loss": 0.7602188587188721,
|
| 989 |
+
"eval_runtime": 1.1738,
|
| 990 |
+
"eval_samples_per_second": 97.117,
|
| 991 |
+
"eval_steps_per_second": 5.111,
|
| 992 |
+
"step": 650
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.0307692307692307,
|
| 996 |
+
"grad_norm": 1.3759773969650269,
|
| 997 |
+
"learning_rate": 1.935965028223259e-05,
|
| 998 |
+
"loss": 0.2182,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.0307692307692307,
|
| 1003 |
+
"eval_loss": 0.8499692678451538,
|
| 1004 |
+
"eval_runtime": 1.1719,
|
| 1005 |
+
"eval_samples_per_second": 97.276,
|
| 1006 |
+
"eval_steps_per_second": 5.12,
|
| 1007 |
+
"step": 660
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.0615384615384613,
|
| 1011 |
+
"grad_norm": 1.4757074117660522,
|
| 1012 |
+
"learning_rate": 1.932129465573568e-05,
|
| 1013 |
+
"loss": 0.2196,
|
| 1014 |
+
"step": 670
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.0615384615384613,
|
| 1018 |
+
"eval_loss": 0.866599977016449,
|
| 1019 |
+
"eval_runtime": 1.1739,
|
| 1020 |
+
"eval_samples_per_second": 97.115,
|
| 1021 |
+
"eval_steps_per_second": 5.111,
|
| 1022 |
+
"step": 670
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.0923076923076924,
|
| 1026 |
+
"grad_norm": 1.2500200271606445,
|
| 1027 |
+
"learning_rate": 1.9281863752957095e-05,
|
| 1028 |
+
"loss": 0.1772,
|
| 1029 |
+
"step": 680
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.0923076923076924,
|
| 1033 |
+
"eval_loss": 0.8642630577087402,
|
| 1034 |
+
"eval_runtime": 1.1753,
|
| 1035 |
+
"eval_samples_per_second": 96.999,
|
| 1036 |
+
"eval_steps_per_second": 5.105,
|
| 1037 |
+
"step": 680
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.123076923076923,
|
| 1041 |
+
"grad_norm": 2.561828851699829,
|
| 1042 |
+
"learning_rate": 1.92413621225262e-05,
|
| 1043 |
+
"loss": 0.2014,
|
| 1044 |
+
"step": 690
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.123076923076923,
|
| 1048 |
+
"eval_loss": 0.8872668743133545,
|
| 1049 |
+
"eval_runtime": 1.1761,
|
| 1050 |
+
"eval_samples_per_second": 96.933,
|
| 1051 |
+
"eval_steps_per_second": 5.102,
|
| 1052 |
+
"step": 690
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.1538461538461537,
|
| 1056 |
+
"grad_norm": 1.8431521654129028,
|
| 1057 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 1058 |
+
"loss": 0.2245,
|
| 1059 |
+
"step": 700
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.1538461538461537,
|
| 1063 |
+
"eval_loss": 0.8799911737442017,
|
| 1064 |
+
"eval_runtime": 1.1766,
|
| 1065 |
+
"eval_samples_per_second": 96.89,
|
| 1066 |
+
"eval_steps_per_second": 5.099,
|
| 1067 |
+
"step": 700
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.184615384615385,
|
| 1071 |
+
"grad_norm": 1.4718531370162964,
|
| 1072 |
+
"learning_rate": 1.915716549026541e-05,
|
| 1073 |
+
"loss": 0.1911,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.184615384615385,
|
| 1078 |
+
"eval_loss": 0.8806982636451721,
|
| 1079 |
+
"eval_runtime": 1.1696,
|
| 1080 |
+
"eval_samples_per_second": 97.473,
|
| 1081 |
+
"eval_steps_per_second": 5.13,
|
| 1082 |
+
"step": 710
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.2153846153846155,
|
| 1086 |
+
"grad_norm": 2.061520576477051,
|
| 1087 |
+
"learning_rate": 1.9113480201103658e-05,
|
| 1088 |
+
"loss": 0.232,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.2153846153846155,
|
| 1093 |
+
"eval_loss": 0.8812103271484375,
|
| 1094 |
+
"eval_runtime": 1.1676,
|
| 1095 |
+
"eval_samples_per_second": 97.636,
|
| 1096 |
+
"eval_steps_per_second": 5.139,
|
| 1097 |
+
"step": 720
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.246153846153846,
|
| 1101 |
+
"grad_norm": 2.530285596847534,
|
| 1102 |
+
"learning_rate": 1.9068743608505454e-05,
|
| 1103 |
+
"loss": 0.2232,
|
| 1104 |
+
"step": 730
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.246153846153846,
|
| 1108 |
+
"eval_loss": 0.8682798147201538,
|
| 1109 |
+
"eval_runtime": 1.1687,
|
| 1110 |
+
"eval_samples_per_second": 97.545,
|
| 1111 |
+
"eval_steps_per_second": 5.134,
|
| 1112 |
+
"step": 730
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.276923076923077,
|
| 1116 |
+
"grad_norm": 1.8433207273483276,
|
| 1117 |
+
"learning_rate": 1.902296087314845e-05,
|
| 1118 |
+
"loss": 0.2315,
|
| 1119 |
+
"step": 740
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.276923076923077,
|
| 1123 |
+
"eval_loss": 0.8747490048408508,
|
| 1124 |
+
"eval_runtime": 1.1702,
|
| 1125 |
+
"eval_samples_per_second": 97.421,
|
| 1126 |
+
"eval_steps_per_second": 5.127,
|
| 1127 |
+
"step": 740
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.3076923076923075,
|
| 1131 |
+
"grad_norm": 2.1529664993286133,
|
| 1132 |
+
"learning_rate": 1.8976137276390145e-05,
|
| 1133 |
+
"loss": 0.239,
|
| 1134 |
+
"step": 750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.3076923076923075,
|
| 1138 |
+
"eval_loss": 0.8778530359268188,
|
| 1139 |
+
"eval_runtime": 1.1674,
|
| 1140 |
+
"eval_samples_per_second": 97.656,
|
| 1141 |
+
"eval_steps_per_second": 5.14,
|
| 1142 |
+
"step": 750
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.3384615384615386,
|
| 1146 |
+
"grad_norm": 1.839979887008667,
|
| 1147 |
+
"learning_rate": 1.892827821965864e-05,
|
| 1148 |
+
"loss": 0.2228,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.3384615384615386,
|
| 1153 |
+
"eval_loss": 0.8769873976707458,
|
| 1154 |
+
"eval_runtime": 1.1718,
|
| 1155 |
+
"eval_samples_per_second": 97.288,
|
| 1156 |
+
"eval_steps_per_second": 5.12,
|
| 1157 |
+
"step": 760
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.3692307692307693,
|
| 1161 |
+
"grad_norm": 1.3814674615859985,
|
| 1162 |
+
"learning_rate": 1.8879389223829592e-05,
|
| 1163 |
+
"loss": 0.1999,
|
| 1164 |
+
"step": 770
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.3692307692307693,
|
| 1168 |
+
"eval_loss": 0.8731983304023743,
|
| 1169 |
+
"eval_runtime": 1.1723,
|
| 1170 |
+
"eval_samples_per_second": 97.242,
|
| 1171 |
+
"eval_steps_per_second": 5.118,
|
| 1172 |
+
"step": 770
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.4,
|
| 1176 |
+
"grad_norm": 1.5382564067840576,
|
| 1177 |
+
"learning_rate": 1.8829475928589272e-05,
|
| 1178 |
+
"loss": 0.1888,
|
| 1179 |
+
"step": 780
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.4,
|
| 1183 |
+
"eval_loss": 0.8794592022895813,
|
| 1184 |
+
"eval_runtime": 1.1678,
|
| 1185 |
+
"eval_samples_per_second": 97.622,
|
| 1186 |
+
"eval_steps_per_second": 5.138,
|
| 1187 |
+
"step": 780
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.430769230769231,
|
| 1191 |
+
"grad_norm": 2.371277093887329,
|
| 1192 |
+
"learning_rate": 1.8778544091784047e-05,
|
| 1193 |
+
"loss": 0.2293,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.430769230769231,
|
| 1198 |
+
"eval_loss": 0.8893064260482788,
|
| 1199 |
+
"eval_runtime": 1.1684,
|
| 1200 |
+
"eval_samples_per_second": 97.567,
|
| 1201 |
+
"eval_steps_per_second": 5.135,
|
| 1202 |
+
"step": 790
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.4615384615384617,
|
| 1206 |
+
"grad_norm": 1.4892629384994507,
|
| 1207 |
+
"learning_rate": 1.8726599588756144e-05,
|
| 1208 |
+
"loss": 0.2326,
|
| 1209 |
+
"step": 800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.4615384615384617,
|
| 1213 |
+
"eval_loss": 0.8676194548606873,
|
| 1214 |
+
"eval_runtime": 1.1699,
|
| 1215 |
+
"eval_samples_per_second": 97.441,
|
| 1216 |
+
"eval_steps_per_second": 5.128,
|
| 1217 |
+
"step": 800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.4923076923076923,
|
| 1221 |
+
"grad_norm": 1.6047152280807495,
|
| 1222 |
+
"learning_rate": 1.8673648411665895e-05,
|
| 1223 |
+
"loss": 0.1986,
|
| 1224 |
+
"step": 810
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.4923076923076923,
|
| 1228 |
+
"eval_loss": 0.8942193984985352,
|
| 1229 |
+
"eval_runtime": 1.17,
|
| 1230 |
+
"eval_samples_per_second": 97.433,
|
| 1231 |
+
"eval_steps_per_second": 5.128,
|
| 1232 |
+
"step": 810
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.523076923076923,
|
| 1236 |
+
"grad_norm": 2.106461524963379,
|
| 1237 |
+
"learning_rate": 1.8619696668800494e-05,
|
| 1238 |
+
"loss": 0.238,
|
| 1239 |
+
"step": 820
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.523076923076923,
|
| 1243 |
+
"eval_loss": 0.8907297849655151,
|
| 1244 |
+
"eval_runtime": 1.1681,
|
| 1245 |
+
"eval_samples_per_second": 97.593,
|
| 1246 |
+
"eval_steps_per_second": 5.136,
|
| 1247 |
+
"step": 820
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 2.5538461538461537,
|
| 1251 |
+
"grad_norm": 1.7706941366195679,
|
| 1252 |
+
"learning_rate": 1.8564750583869374e-05,
|
| 1253 |
+
"loss": 0.2278,
|
| 1254 |
+
"step": 830
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 2.5538461538461537,
|
| 1258 |
+
"eval_loss": 0.8588199019432068,
|
| 1259 |
+
"eval_runtime": 1.1709,
|
| 1260 |
+
"eval_samples_per_second": 97.36,
|
| 1261 |
+
"eval_steps_per_second": 5.124,
|
| 1262 |
+
"step": 830
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.5846153846153848,
|
| 1266 |
+
"grad_norm": 1.4446183443069458,
|
| 1267 |
+
"learning_rate": 1.850881649528625e-05,
|
| 1268 |
+
"loss": 0.2033,
|
| 1269 |
+
"step": 840
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.5846153846153848,
|
| 1273 |
+
"eval_loss": 0.8720155954360962,
|
| 1274 |
+
"eval_runtime": 1.1716,
|
| 1275 |
+
"eval_samples_per_second": 97.304,
|
| 1276 |
+
"eval_steps_per_second": 5.121,
|
| 1277 |
+
"step": 840
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.6153846153846154,
|
| 1281 |
+
"grad_norm": 1.5155885219573975,
|
| 1282 |
+
"learning_rate": 1.845190085543795e-05,
|
| 1283 |
+
"loss": 0.2211,
|
| 1284 |
+
"step": 850
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6153846153846154,
|
| 1288 |
+
"eval_loss": 0.8732415437698364,
|
| 1289 |
+
"eval_runtime": 1.1708,
|
| 1290 |
+
"eval_samples_per_second": 97.371,
|
| 1291 |
+
"eval_steps_per_second": 5.125,
|
| 1292 |
+
"step": 850
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.646153846153846,
|
| 1296 |
+
"grad_norm": 2.934957981109619,
|
| 1297 |
+
"learning_rate": 1.839401022994006e-05,
|
| 1298 |
+
"loss": 0.21,
|
| 1299 |
+
"step": 860
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.646153846153846,
|
| 1303 |
+
"eval_loss": 0.8790932297706604,
|
| 1304 |
+
"eval_runtime": 1.1671,
|
| 1305 |
+
"eval_samples_per_second": 97.68,
|
| 1306 |
+
"eval_steps_per_second": 5.141,
|
| 1307 |
+
"step": 860
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.676923076923077,
|
| 1311 |
+
"grad_norm": 2.930457353591919,
|
| 1312 |
+
"learning_rate": 1.8335151296879576e-05,
|
| 1313 |
+
"loss": 0.2313,
|
| 1314 |
+
"step": 870
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.676923076923077,
|
| 1318 |
+
"eval_loss": 0.8555310368537903,
|
| 1319 |
+
"eval_runtime": 1.1718,
|
| 1320 |
+
"eval_samples_per_second": 97.284,
|
| 1321 |
+
"eval_steps_per_second": 5.12,
|
| 1322 |
+
"step": 870
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.707692307692308,
|
| 1326 |
+
"grad_norm": 2.029303789138794,
|
| 1327 |
+
"learning_rate": 1.82753308460445e-05,
|
| 1328 |
+
"loss": 0.2285,
|
| 1329 |
+
"step": 880
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.707692307692308,
|
| 1333 |
+
"eval_loss": 0.870873749256134,
|
| 1334 |
+
"eval_runtime": 1.1696,
|
| 1335 |
+
"eval_samples_per_second": 97.47,
|
| 1336 |
+
"eval_steps_per_second": 5.13,
|
| 1337 |
+
"step": 880
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 2.7384615384615385,
|
| 1341 |
+
"grad_norm": 1.949743390083313,
|
| 1342 |
+
"learning_rate": 1.821455577814062e-05,
|
| 1343 |
+
"loss": 0.2134,
|
| 1344 |
+
"step": 890
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.7384615384615385,
|
| 1348 |
+
"eval_loss": 0.8873838186264038,
|
| 1349 |
+
"eval_runtime": 1.1688,
|
| 1350 |
+
"eval_samples_per_second": 97.538,
|
| 1351 |
+
"eval_steps_per_second": 5.134,
|
| 1352 |
+
"step": 890
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.769230769230769,
|
| 1356 |
+
"grad_norm": 1.6945637464523315,
|
| 1357 |
+
"learning_rate": 1.8152833103995443e-05,
|
| 1358 |
+
"loss": 0.218,
|
| 1359 |
+
"step": 900
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.769230769230769,
|
| 1363 |
+
"eval_loss": 0.8798539042472839,
|
| 1364 |
+
"eval_runtime": 1.1699,
|
| 1365 |
+
"eval_samples_per_second": 97.444,
|
| 1366 |
+
"eval_steps_per_second": 5.129,
|
| 1367 |
+
"step": 900
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8,
|
| 1371 |
+
"grad_norm": 2.2784035205841064,
|
| 1372 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 1373 |
+
"loss": 0.2426,
|
| 1374 |
+
"step": 910
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8,
|
| 1378 |
+
"eval_loss": 0.8753871321678162,
|
| 1379 |
+
"eval_runtime": 1.1723,
|
| 1380 |
+
"eval_samples_per_second": 97.246,
|
| 1381 |
+
"eval_steps_per_second": 5.118,
|
| 1382 |
+
"step": 910
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.830769230769231,
|
| 1386 |
+
"grad_norm": 1.750718593597412,
|
| 1387 |
+
"learning_rate": 1.802657352603483e-05,
|
| 1388 |
+
"loss": 0.2059,
|
| 1389 |
+
"step": 920
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.830769230769231,
|
| 1393 |
+
"eval_loss": 0.8761821985244751,
|
| 1394 |
+
"eval_runtime": 1.1762,
|
| 1395 |
+
"eval_samples_per_second": 96.92,
|
| 1396 |
+
"eval_steps_per_second": 5.101,
|
| 1397 |
+
"step": 920
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.8615384615384616,
|
| 1401 |
+
"grad_norm": 1.2204371690750122,
|
| 1402 |
+
"learning_rate": 1.7962051187141377e-05,
|
| 1403 |
+
"loss": 0.2229,
|
| 1404 |
+
"step": 930
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.8615384615384616,
|
| 1408 |
+
"eval_loss": 0.8772630095481873,
|
| 1409 |
+
"eval_runtime": 1.1697,
|
| 1410 |
+
"eval_samples_per_second": 97.458,
|
| 1411 |
+
"eval_steps_per_second": 5.129,
|
| 1412 |
+
"step": 930
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.8923076923076922,
|
| 1416 |
+
"grad_norm": 2.0748281478881836,
|
| 1417 |
+
"learning_rate": 1.7896610370170452e-05,
|
| 1418 |
+
"loss": 0.2582,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.8923076923076922,
|
| 1423 |
+
"eval_loss": 0.884286105632782,
|
| 1424 |
+
"eval_runtime": 1.1773,
|
| 1425 |
+
"eval_samples_per_second": 96.829,
|
| 1426 |
+
"eval_steps_per_second": 5.096,
|
| 1427 |
+
"step": 940
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.9230769230769234,
|
| 1431 |
+
"grad_norm": 1.9051027297973633,
|
| 1432 |
+
"learning_rate": 1.7830258624176224e-05,
|
| 1433 |
+
"loss": 0.2102,
|
| 1434 |
+
"step": 950
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.9230769230769234,
|
| 1438 |
+
"eval_loss": 0.878767192363739,
|
| 1439 |
+
"eval_runtime": 1.1684,
|
| 1440 |
+
"eval_samples_per_second": 97.571,
|
| 1441 |
+
"eval_steps_per_second": 5.135,
|
| 1442 |
+
"step": 950
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.953846153846154,
|
| 1446 |
+
"grad_norm": 1.8921849727630615,
|
| 1447 |
+
"learning_rate": 1.776300360329488e-05,
|
| 1448 |
+
"loss": 0.2331,
|
| 1449 |
+
"step": 960
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.953846153846154,
|
| 1453 |
+
"eval_loss": 0.8681402206420898,
|
| 1454 |
+
"eval_runtime": 1.1693,
|
| 1455 |
+
"eval_samples_per_second": 97.497,
|
| 1456 |
+
"eval_steps_per_second": 5.131,
|
| 1457 |
+
"step": 960
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.9846153846153847,
|
| 1461 |
+
"grad_norm": 1.5593761205673218,
|
| 1462 |
+
"learning_rate": 1.769485306586166e-05,
|
| 1463 |
+
"loss": 0.2226,
|
| 1464 |
+
"step": 970
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.9846153846153847,
|
| 1468 |
+
"eval_loss": 0.8726215958595276,
|
| 1469 |
+
"eval_runtime": 1.1712,
|
| 1470 |
+
"eval_samples_per_second": 97.336,
|
| 1471 |
+
"eval_steps_per_second": 5.123,
|
| 1472 |
+
"step": 970
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.0153846153846153,
|
| 1476 |
+
"grad_norm": 1.4075913429260254,
|
| 1477 |
+
"learning_rate": 1.762581487351587e-05,
|
| 1478 |
+
"loss": 0.1535,
|
| 1479 |
+
"step": 980
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.0153846153846153,
|
| 1483 |
+
"eval_loss": 0.9113463163375854,
|
| 1484 |
+
"eval_runtime": 1.1794,
|
| 1485 |
+
"eval_samples_per_second": 96.661,
|
| 1486 |
+
"eval_steps_per_second": 5.087,
|
| 1487 |
+
"step": 980
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.046153846153846,
|
| 1491 |
+
"grad_norm": 1.410906434059143,
|
| 1492 |
+
"learning_rate": 1.7555896990294003e-05,
|
| 1493 |
+
"loss": 0.1021,
|
| 1494 |
+
"step": 990
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.046153846153846,
|
| 1498 |
+
"eval_loss": 0.9633604288101196,
|
| 1499 |
+
"eval_runtime": 1.1702,
|
| 1500 |
+
"eval_samples_per_second": 97.416,
|
| 1501 |
+
"eval_steps_per_second": 5.127,
|
| 1502 |
+
"step": 990
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 3.076923076923077,
|
| 1506 |
+
"grad_norm": 1.0470328330993652,
|
| 1507 |
+
"learning_rate": 1.7485107481711014e-05,
|
| 1508 |
+
"loss": 0.1118,
|
| 1509 |
+
"step": 1000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 3.076923076923077,
|
| 1513 |
+
"eval_loss": 0.9647616744041443,
|
| 1514 |
+
"eval_runtime": 1.1683,
|
| 1515 |
+
"eval_samples_per_second": 97.574,
|
| 1516 |
+
"eval_steps_per_second": 5.135,
|
| 1517 |
+
"step": 1000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.1076923076923078,
|
| 1521 |
+
"grad_norm": 2.098592758178711,
|
| 1522 |
+
"learning_rate": 1.741345451382992e-05,
|
| 1523 |
+
"loss": 0.1224,
|
| 1524 |
+
"step": 1010
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.1076923076923078,
|
| 1528 |
+
"eval_loss": 0.9478728175163269,
|
| 1529 |
+
"eval_runtime": 1.1735,
|
| 1530 |
+
"eval_samples_per_second": 97.143,
|
| 1531 |
+
"eval_steps_per_second": 5.113,
|
| 1532 |
+
"step": 1010
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 3.1384615384615384,
|
| 1536 |
+
"grad_norm": 1.6263700723648071,
|
| 1537 |
+
"learning_rate": 1.7340946352319795e-05,
|
| 1538 |
+
"loss": 0.1377,
|
| 1539 |
+
"step": 1020
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 3.1384615384615384,
|
| 1543 |
+
"eval_loss": 0.922594428062439,
|
| 1544 |
+
"eval_runtime": 1.1721,
|
| 1545 |
+
"eval_samples_per_second": 97.262,
|
| 1546 |
+
"eval_steps_per_second": 5.119,
|
| 1547 |
+
"step": 1020
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 3.169230769230769,
|
| 1551 |
+
"grad_norm": 1.0413360595703125,
|
| 1552 |
+
"learning_rate": 1.7267591361502233e-05,
|
| 1553 |
+
"loss": 0.1249,
|
| 1554 |
+
"step": 1030
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 3.169230769230769,
|
| 1558 |
+
"eval_loss": 0.9370183944702148,
|
| 1559 |
+
"eval_runtime": 1.169,
|
| 1560 |
+
"eval_samples_per_second": 97.519,
|
| 1561 |
+
"eval_steps_per_second": 5.133,
|
| 1562 |
+
"step": 1030
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 3.2,
|
| 1566 |
+
"grad_norm": 1.6745877265930176,
|
| 1567 |
+
"learning_rate": 1.7193398003386514e-05,
|
| 1568 |
+
"loss": 0.1186,
|
| 1569 |
+
"step": 1040
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 3.2,
|
| 1573 |
+
"eval_loss": 0.9573487639427185,
|
| 1574 |
+
"eval_runtime": 1.1681,
|
| 1575 |
+
"eval_samples_per_second": 97.596,
|
| 1576 |
+
"eval_steps_per_second": 5.137,
|
| 1577 |
+
"step": 1040
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.230769230769231,
|
| 1581 |
+
"grad_norm": 1.3979861736297607,
|
| 1582 |
+
"learning_rate": 1.7118374836693407e-05,
|
| 1583 |
+
"loss": 0.135,
|
| 1584 |
+
"step": 1050
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.230769230769231,
|
| 1588 |
+
"eval_loss": 0.9554935693740845,
|
| 1589 |
+
"eval_runtime": 1.1664,
|
| 1590 |
+
"eval_samples_per_second": 97.735,
|
| 1591 |
+
"eval_steps_per_second": 5.144,
|
| 1592 |
+
"step": 1050
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 3.2615384615384615,
|
| 1596 |
+
"grad_norm": 1.4017126560211182,
|
| 1597 |
+
"learning_rate": 1.7042530515867897e-05,
|
| 1598 |
+
"loss": 0.1102,
|
| 1599 |
+
"step": 1060
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 3.2615384615384615,
|
| 1603 |
+
"eval_loss": 0.9547311663627625,
|
| 1604 |
+
"eval_runtime": 1.1685,
|
| 1605 |
+
"eval_samples_per_second": 97.561,
|
| 1606 |
+
"eval_steps_per_second": 5.135,
|
| 1607 |
+
"step": 1060
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 3.292307692307692,
|
| 1611 |
+
"grad_norm": 1.3390568494796753,
|
| 1612 |
+
"learning_rate": 1.6965873790080806e-05,
|
| 1613 |
+
"loss": 0.1358,
|
| 1614 |
+
"step": 1070
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 3.292307692307692,
|
| 1618 |
+
"eval_loss": 0.9695214629173279,
|
| 1619 |
+
"eval_runtime": 1.1695,
|
| 1620 |
+
"eval_samples_per_second": 97.474,
|
| 1621 |
+
"eval_steps_per_second": 5.13,
|
| 1622 |
+
"step": 1070
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 3.3230769230769233,
|
| 1626 |
+
"grad_norm": 1.9002848863601685,
|
| 1627 |
+
"learning_rate": 1.6888413502219534e-05,
|
| 1628 |
+
"loss": 0.1419,
|
| 1629 |
+
"step": 1080
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 3.3230769230769233,
|
| 1633 |
+
"eval_loss": 0.9528248310089111,
|
| 1634 |
+
"eval_runtime": 1.174,
|
| 1635 |
+
"eval_samples_per_second": 97.107,
|
| 1636 |
+
"eval_steps_per_second": 5.111,
|
| 1637 |
+
"step": 1080
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 3.353846153846154,
|
| 1641 |
+
"grad_norm": 1.7857142686843872,
|
| 1642 |
+
"learning_rate": 1.6810158587867973e-05,
|
| 1643 |
+
"loss": 0.128,
|
| 1644 |
+
"step": 1090
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 3.353846153846154,
|
| 1648 |
+
"eval_loss": 0.9331367611885071,
|
| 1649 |
+
"eval_runtime": 1.1718,
|
| 1650 |
+
"eval_samples_per_second": 97.282,
|
| 1651 |
+
"eval_steps_per_second": 5.12,
|
| 1652 |
+
"step": 1090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 3.3846153846153846,
|
| 1656 |
+
"grad_norm": 1.5178117752075195,
|
| 1657 |
+
"learning_rate": 1.67311180742757e-05,
|
| 1658 |
+
"loss": 0.112,
|
| 1659 |
+
"step": 1100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 3.3846153846153846,
|
| 1663 |
+
"eval_loss": 0.9547919631004333,
|
| 1664 |
+
"eval_runtime": 1.1695,
|
| 1665 |
+
"eval_samples_per_second": 97.48,
|
| 1666 |
+
"eval_steps_per_second": 5.131,
|
| 1667 |
+
"step": 1100
|
| 1668 |
+
}
|
| 1669 |
+
],
|
| 1670 |
+
"logging_steps": 10,
|
| 1671 |
+
"max_steps": 3250,
|
| 1672 |
+
"num_input_tokens_seen": 0,
|
| 1673 |
+
"num_train_epochs": 10,
|
| 1674 |
+
"save_steps": 50,
|
| 1675 |
+
"total_flos": 6.860504988175565e+16,
|
| 1676 |
+
"train_batch_size": 2,
|
| 1677 |
+
"trial_name": null,
|
| 1678 |
+
"trial_params": null
|
| 1679 |
+
}
|
{checkpoint-200 → checkpoint-1100}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/added_tokens.json
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/config.json
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/generation_config.json
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/model.safetensors.index.json
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/special_tokens_map.json
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/tokenizer.json
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/tokenizer.model
RENAMED
|
File without changes
|
{checkpoint-250 → checkpoint-1150}/tokenizer_config.json
RENAMED
|
File without changes
|
checkpoint-1150/trainer_state.json
ADDED
|
@@ -0,0 +1,1754 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.5384615384615383,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1150,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006153846153846154,
|
| 13 |
+
"eval_loss": 1.487821340560913,
|
| 14 |
+
"eval_runtime": 1.2374,
|
| 15 |
+
"eval_samples_per_second": 92.131,
|
| 16 |
+
"eval_steps_per_second": 4.849,
|
| 17 |
+
"step": 2
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03076923076923077,
|
| 21 |
+
"grad_norm": 20.007999420166016,
|
| 22 |
+
"learning_rate": 6.153846153846155e-07,
|
| 23 |
+
"loss": 1.4703,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03076923076923077,
|
| 28 |
+
"eval_loss": 1.4221965074539185,
|
| 29 |
+
"eval_runtime": 1.1701,
|
| 30 |
+
"eval_samples_per_second": 97.431,
|
| 31 |
+
"eval_steps_per_second": 5.128,
|
| 32 |
+
"step": 10
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06153846153846154,
|
| 36 |
+
"grad_norm": 29.657033920288086,
|
| 37 |
+
"learning_rate": 1.230769230769231e-06,
|
| 38 |
+
"loss": 1.4402,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06153846153846154,
|
| 43 |
+
"eval_loss": 1.238189935684204,
|
| 44 |
+
"eval_runtime": 1.1729,
|
| 45 |
+
"eval_samples_per_second": 97.195,
|
| 46 |
+
"eval_steps_per_second": 5.116,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.09230769230769231,
|
| 51 |
+
"grad_norm": 10.80688190460205,
|
| 52 |
+
"learning_rate": 1.8461538461538465e-06,
|
| 53 |
+
"loss": 1.0957,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09230769230769231,
|
| 58 |
+
"eval_loss": 0.908875048160553,
|
| 59 |
+
"eval_runtime": 1.1739,
|
| 60 |
+
"eval_samples_per_second": 97.111,
|
| 61 |
+
"eval_steps_per_second": 5.111,
|
| 62 |
+
"step": 30
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.12307692307692308,
|
| 66 |
+
"grad_norm": 12.914667129516602,
|
| 67 |
+
"learning_rate": 2.461538461538462e-06,
|
| 68 |
+
"loss": 0.8136,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.12307692307692308,
|
| 73 |
+
"eval_loss": 0.7436326146125793,
|
| 74 |
+
"eval_runtime": 1.1712,
|
| 75 |
+
"eval_samples_per_second": 97.34,
|
| 76 |
+
"eval_steps_per_second": 5.123,
|
| 77 |
+
"step": 40
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.15384615384615385,
|
| 81 |
+
"grad_norm": 5.206438064575195,
|
| 82 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 83 |
+
"loss": 0.7864,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.15384615384615385,
|
| 88 |
+
"eval_loss": 0.6924600601196289,
|
| 89 |
+
"eval_runtime": 1.1718,
|
| 90 |
+
"eval_samples_per_second": 97.287,
|
| 91 |
+
"eval_steps_per_second": 5.12,
|
| 92 |
+
"step": 50
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.18461538461538463,
|
| 96 |
+
"grad_norm": 5.670599460601807,
|
| 97 |
+
"learning_rate": 3.692307692307693e-06,
|
| 98 |
+
"loss": 0.622,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.18461538461538463,
|
| 103 |
+
"eval_loss": 0.6791077852249146,
|
| 104 |
+
"eval_runtime": 1.1678,
|
| 105 |
+
"eval_samples_per_second": 97.62,
|
| 106 |
+
"eval_steps_per_second": 5.138,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.2153846153846154,
|
| 111 |
+
"grad_norm": 3.6266868114471436,
|
| 112 |
+
"learning_rate": 4.307692307692308e-06,
|
| 113 |
+
"loss": 0.6672,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2153846153846154,
|
| 118 |
+
"eval_loss": 0.6631948947906494,
|
| 119 |
+
"eval_runtime": 1.1626,
|
| 120 |
+
"eval_samples_per_second": 98.06,
|
| 121 |
+
"eval_steps_per_second": 5.161,
|
| 122 |
+
"step": 70
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.24615384615384617,
|
| 126 |
+
"grad_norm": 3.1027884483337402,
|
| 127 |
+
"learning_rate": 4.923076923076924e-06,
|
| 128 |
+
"loss": 0.6097,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.24615384615384617,
|
| 133 |
+
"eval_loss": 0.6625475883483887,
|
| 134 |
+
"eval_runtime": 1.1717,
|
| 135 |
+
"eval_samples_per_second": 97.294,
|
| 136 |
+
"eval_steps_per_second": 5.121,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.27692307692307694,
|
| 141 |
+
"grad_norm": 3.233839511871338,
|
| 142 |
+
"learning_rate": 5.538461538461539e-06,
|
| 143 |
+
"loss": 0.6957,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.27692307692307694,
|
| 148 |
+
"eval_loss": 0.6590337157249451,
|
| 149 |
+
"eval_runtime": 1.1664,
|
| 150 |
+
"eval_samples_per_second": 97.738,
|
| 151 |
+
"eval_steps_per_second": 5.144,
|
| 152 |
+
"step": 90
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.3076923076923077,
|
| 156 |
+
"grad_norm": 4.932614803314209,
|
| 157 |
+
"learning_rate": 6.153846153846155e-06,
|
| 158 |
+
"loss": 0.6994,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.3076923076923077,
|
| 163 |
+
"eval_loss": 0.6592673659324646,
|
| 164 |
+
"eval_runtime": 1.1692,
|
| 165 |
+
"eval_samples_per_second": 97.502,
|
| 166 |
+
"eval_steps_per_second": 5.132,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.3384615384615385,
|
| 171 |
+
"grad_norm": 2.642303466796875,
|
| 172 |
+
"learning_rate": 6.76923076923077e-06,
|
| 173 |
+
"loss": 0.6501,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.3384615384615385,
|
| 178 |
+
"eval_loss": 0.6589992642402649,
|
| 179 |
+
"eval_runtime": 1.1669,
|
| 180 |
+
"eval_samples_per_second": 97.691,
|
| 181 |
+
"eval_steps_per_second": 5.142,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.36923076923076925,
|
| 186 |
+
"grad_norm": 3.45409893989563,
|
| 187 |
+
"learning_rate": 7.384615384615386e-06,
|
| 188 |
+
"loss": 0.7512,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.36923076923076925,
|
| 193 |
+
"eval_loss": 0.6568591594696045,
|
| 194 |
+
"eval_runtime": 1.1778,
|
| 195 |
+
"eval_samples_per_second": 96.79,
|
| 196 |
+
"eval_steps_per_second": 5.094,
|
| 197 |
+
"step": 120
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.4,
|
| 201 |
+
"grad_norm": 8.460351943969727,
|
| 202 |
+
"learning_rate": 8.000000000000001e-06,
|
| 203 |
+
"loss": 0.5996,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4,
|
| 208 |
+
"eval_loss": 0.6578297019004822,
|
| 209 |
+
"eval_runtime": 1.1683,
|
| 210 |
+
"eval_samples_per_second": 97.575,
|
| 211 |
+
"eval_steps_per_second": 5.136,
|
| 212 |
+
"step": 130
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4307692307692308,
|
| 216 |
+
"grad_norm": 2.4495465755462646,
|
| 217 |
+
"learning_rate": 8.615384615384617e-06,
|
| 218 |
+
"loss": 0.6498,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4307692307692308,
|
| 223 |
+
"eval_loss": 0.6588199138641357,
|
| 224 |
+
"eval_runtime": 1.1664,
|
| 225 |
+
"eval_samples_per_second": 97.739,
|
| 226 |
+
"eval_steps_per_second": 5.144,
|
| 227 |
+
"step": 140
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.46153846153846156,
|
| 231 |
+
"grad_norm": 2.6779162883758545,
|
| 232 |
+
"learning_rate": 9.230769230769232e-06,
|
| 233 |
+
"loss": 0.703,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.46153846153846156,
|
| 238 |
+
"eval_loss": 0.6626676321029663,
|
| 239 |
+
"eval_runtime": 1.1764,
|
| 240 |
+
"eval_samples_per_second": 96.904,
|
| 241 |
+
"eval_steps_per_second": 5.1,
|
| 242 |
+
"step": 150
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.49230769230769234,
|
| 246 |
+
"grad_norm": 8.147880554199219,
|
| 247 |
+
"learning_rate": 9.846153846153848e-06,
|
| 248 |
+
"loss": 0.6649,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.49230769230769234,
|
| 253 |
+
"eval_loss": 0.6644802689552307,
|
| 254 |
+
"eval_runtime": 1.1701,
|
| 255 |
+
"eval_samples_per_second": 97.424,
|
| 256 |
+
"eval_steps_per_second": 5.128,
|
| 257 |
+
"step": 160
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.5230769230769231,
|
| 261 |
+
"grad_norm": 2.7410728931427,
|
| 262 |
+
"learning_rate": 1.0461538461538463e-05,
|
| 263 |
+
"loss": 0.6704,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.5230769230769231,
|
| 268 |
+
"eval_loss": 0.6591455340385437,
|
| 269 |
+
"eval_runtime": 1.1722,
|
| 270 |
+
"eval_samples_per_second": 97.253,
|
| 271 |
+
"eval_steps_per_second": 5.119,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.5538461538461539,
|
| 276 |
+
"grad_norm": 6.886598587036133,
|
| 277 |
+
"learning_rate": 1.1076923076923079e-05,
|
| 278 |
+
"loss": 0.6073,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.5538461538461539,
|
| 283 |
+
"eval_loss": 0.6617754697799683,
|
| 284 |
+
"eval_runtime": 1.1682,
|
| 285 |
+
"eval_samples_per_second": 97.588,
|
| 286 |
+
"eval_steps_per_second": 5.136,
|
| 287 |
+
"step": 180
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.5846153846153846,
|
| 291 |
+
"grad_norm": 4.26249361038208,
|
| 292 |
+
"learning_rate": 1.1692307692307694e-05,
|
| 293 |
+
"loss": 0.7556,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.5846153846153846,
|
| 298 |
+
"eval_loss": 0.6659030318260193,
|
| 299 |
+
"eval_runtime": 1.1721,
|
| 300 |
+
"eval_samples_per_second": 97.265,
|
| 301 |
+
"eval_steps_per_second": 5.119,
|
| 302 |
+
"step": 190
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.6153846153846154,
|
| 306 |
+
"grad_norm": 4.204222202301025,
|
| 307 |
+
"learning_rate": 1.230769230769231e-05,
|
| 308 |
+
"loss": 0.6583,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.6153846153846154,
|
| 313 |
+
"eval_loss": 0.6660778522491455,
|
| 314 |
+
"eval_runtime": 1.1728,
|
| 315 |
+
"eval_samples_per_second": 97.201,
|
| 316 |
+
"eval_steps_per_second": 5.116,
|
| 317 |
+
"step": 200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6461538461538462,
|
| 321 |
+
"grad_norm": 3.154658079147339,
|
| 322 |
+
"learning_rate": 1.2923076923076925e-05,
|
| 323 |
+
"loss": 0.7288,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6461538461538462,
|
| 328 |
+
"eval_loss": 0.6642422676086426,
|
| 329 |
+
"eval_runtime": 1.166,
|
| 330 |
+
"eval_samples_per_second": 97.772,
|
| 331 |
+
"eval_steps_per_second": 5.146,
|
| 332 |
+
"step": 210
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.676923076923077,
|
| 336 |
+
"grad_norm": 3.248359203338623,
|
| 337 |
+
"learning_rate": 1.353846153846154e-05,
|
| 338 |
+
"loss": 0.6702,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.676923076923077,
|
| 343 |
+
"eval_loss": 0.6680281758308411,
|
| 344 |
+
"eval_runtime": 1.1691,
|
| 345 |
+
"eval_samples_per_second": 97.507,
|
| 346 |
+
"eval_steps_per_second": 5.132,
|
| 347 |
+
"step": 220
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.7076923076923077,
|
| 351 |
+
"grad_norm": 2.5507216453552246,
|
| 352 |
+
"learning_rate": 1.4153846153846156e-05,
|
| 353 |
+
"loss": 0.66,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.7076923076923077,
|
| 358 |
+
"eval_loss": 0.6713310480117798,
|
| 359 |
+
"eval_runtime": 1.167,
|
| 360 |
+
"eval_samples_per_second": 97.683,
|
| 361 |
+
"eval_steps_per_second": 5.141,
|
| 362 |
+
"step": 230
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.7384615384615385,
|
| 366 |
+
"grad_norm": 2.1180007457733154,
|
| 367 |
+
"learning_rate": 1.4769230769230772e-05,
|
| 368 |
+
"loss": 0.6172,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.7384615384615385,
|
| 373 |
+
"eval_loss": 0.6759226322174072,
|
| 374 |
+
"eval_runtime": 1.1758,
|
| 375 |
+
"eval_samples_per_second": 96.954,
|
| 376 |
+
"eval_steps_per_second": 5.103,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.7692307692307693,
|
| 381 |
+
"grad_norm": 2.8318371772766113,
|
| 382 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 383 |
+
"loss": 0.5954,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.7692307692307693,
|
| 388 |
+
"eval_loss": 0.6854314208030701,
|
| 389 |
+
"eval_runtime": 1.1698,
|
| 390 |
+
"eval_samples_per_second": 97.457,
|
| 391 |
+
"eval_steps_per_second": 5.129,
|
| 392 |
+
"step": 250
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.8,
|
| 396 |
+
"grad_norm": 3.519829034805298,
|
| 397 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 398 |
+
"loss": 0.6639,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.8,
|
| 403 |
+
"eval_loss": 0.6849851012229919,
|
| 404 |
+
"eval_runtime": 1.1732,
|
| 405 |
+
"eval_samples_per_second": 97.173,
|
| 406 |
+
"eval_steps_per_second": 5.114,
|
| 407 |
+
"step": 260
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.8307692307692308,
|
| 411 |
+
"grad_norm": 3.7909388542175293,
|
| 412 |
+
"learning_rate": 1.6615384615384618e-05,
|
| 413 |
+
"loss": 0.6892,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.8307692307692308,
|
| 418 |
+
"eval_loss": 0.6863855123519897,
|
| 419 |
+
"eval_runtime": 1.1689,
|
| 420 |
+
"eval_samples_per_second": 97.53,
|
| 421 |
+
"eval_steps_per_second": 5.133,
|
| 422 |
+
"step": 270
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8615384615384616,
|
| 426 |
+
"grad_norm": 8.278326988220215,
|
| 427 |
+
"learning_rate": 1.7230769230769234e-05,
|
| 428 |
+
"loss": 0.6014,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8615384615384616,
|
| 433 |
+
"eval_loss": 0.6890373826026917,
|
| 434 |
+
"eval_runtime": 1.1722,
|
| 435 |
+
"eval_samples_per_second": 97.251,
|
| 436 |
+
"eval_steps_per_second": 5.118,
|
| 437 |
+
"step": 280
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8923076923076924,
|
| 441 |
+
"grad_norm": 2.616001605987549,
|
| 442 |
+
"learning_rate": 1.784615384615385e-05,
|
| 443 |
+
"loss": 0.6775,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.8923076923076924,
|
| 448 |
+
"eval_loss": 0.6797110438346863,
|
| 449 |
+
"eval_runtime": 1.1888,
|
| 450 |
+
"eval_samples_per_second": 95.897,
|
| 451 |
+
"eval_steps_per_second": 5.047,
|
| 452 |
+
"step": 290
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.9230769230769231,
|
| 456 |
+
"grad_norm": 2.4450528621673584,
|
| 457 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 458 |
+
"loss": 0.7434,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.9230769230769231,
|
| 463 |
+
"eval_loss": 0.6779448986053467,
|
| 464 |
+
"eval_runtime": 1.1722,
|
| 465 |
+
"eval_samples_per_second": 97.254,
|
| 466 |
+
"eval_steps_per_second": 5.119,
|
| 467 |
+
"step": 300
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.9538461538461539,
|
| 471 |
+
"grad_norm": 4.121164798736572,
|
| 472 |
+
"learning_rate": 1.907692307692308e-05,
|
| 473 |
+
"loss": 0.6413,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.9538461538461539,
|
| 478 |
+
"eval_loss": 0.6819297671318054,
|
| 479 |
+
"eval_runtime": 1.1683,
|
| 480 |
+
"eval_samples_per_second": 97.574,
|
| 481 |
+
"eval_steps_per_second": 5.135,
|
| 482 |
+
"step": 310
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.9846153846153847,
|
| 486 |
+
"grad_norm": 2.057229995727539,
|
| 487 |
+
"learning_rate": 1.9692307692307696e-05,
|
| 488 |
+
"loss": 0.6511,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.9846153846153847,
|
| 493 |
+
"eval_loss": 0.6861481666564941,
|
| 494 |
+
"eval_runtime": 1.1699,
|
| 495 |
+
"eval_samples_per_second": 97.444,
|
| 496 |
+
"eval_steps_per_second": 5.129,
|
| 497 |
+
"step": 320
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.0153846153846153,
|
| 501 |
+
"grad_norm": 1.7295210361480713,
|
| 502 |
+
"learning_rate": 1.9999855802751384e-05,
|
| 503 |
+
"loss": 0.5872,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.0153846153846153,
|
| 508 |
+
"eval_loss": 0.7481781244277954,
|
| 509 |
+
"eval_runtime": 1.172,
|
| 510 |
+
"eval_samples_per_second": 97.273,
|
| 511 |
+
"eval_steps_per_second": 5.12,
|
| 512 |
+
"step": 330
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.0461538461538462,
|
| 516 |
+
"grad_norm": 3.6910336017608643,
|
| 517 |
+
"learning_rate": 1.9998702249713747e-05,
|
| 518 |
+
"loss": 0.438,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.0461538461538462,
|
| 523 |
+
"eval_loss": 0.7471169829368591,
|
| 524 |
+
"eval_runtime": 1.175,
|
| 525 |
+
"eval_samples_per_second": 97.019,
|
| 526 |
+
"eval_steps_per_second": 5.106,
|
| 527 |
+
"step": 340
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.0769230769230769,
|
| 531 |
+
"grad_norm": 3.9220831394195557,
|
| 532 |
+
"learning_rate": 1.9996395276708856e-05,
|
| 533 |
+
"loss": 0.4141,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0769230769230769,
|
| 538 |
+
"eval_loss": 0.7425897121429443,
|
| 539 |
+
"eval_runtime": 1.1888,
|
| 540 |
+
"eval_samples_per_second": 95.893,
|
| 541 |
+
"eval_steps_per_second": 5.047,
|
| 542 |
+
"step": 350
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.1076923076923078,
|
| 546 |
+
"grad_norm": 2.1136879920959473,
|
| 547 |
+
"learning_rate": 1.9992935149862116e-05,
|
| 548 |
+
"loss": 0.4173,
|
| 549 |
+
"step": 360
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1076923076923078,
|
| 553 |
+
"eval_loss": 0.7477542161941528,
|
| 554 |
+
"eval_runtime": 1.1705,
|
| 555 |
+
"eval_samples_per_second": 97.398,
|
| 556 |
+
"eval_steps_per_second": 5.126,
|
| 557 |
+
"step": 360
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.1384615384615384,
|
| 561 |
+
"grad_norm": 2.187299966812134,
|
| 562 |
+
"learning_rate": 1.998832226832327e-05,
|
| 563 |
+
"loss": 0.4242,
|
| 564 |
+
"step": 370
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.1384615384615384,
|
| 568 |
+
"eval_loss": 0.7475994229316711,
|
| 569 |
+
"eval_runtime": 1.1683,
|
| 570 |
+
"eval_samples_per_second": 97.582,
|
| 571 |
+
"eval_steps_per_second": 5.136,
|
| 572 |
+
"step": 370
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.1692307692307693,
|
| 576 |
+
"grad_norm": 2.683750629425049,
|
| 577 |
+
"learning_rate": 1.9982557164220335e-05,
|
| 578 |
+
"loss": 0.4413,
|
| 579 |
+
"step": 380
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.1692307692307693,
|
| 583 |
+
"eval_loss": 0.7552060484886169,
|
| 584 |
+
"eval_runtime": 1.1689,
|
| 585 |
+
"eval_samples_per_second": 97.529,
|
| 586 |
+
"eval_steps_per_second": 5.133,
|
| 587 |
+
"step": 380
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.2,
|
| 591 |
+
"grad_norm": 3.312657117843628,
|
| 592 |
+
"learning_rate": 1.9975640502598243e-05,
|
| 593 |
+
"loss": 0.3361,
|
| 594 |
+
"step": 390
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.2,
|
| 598 |
+
"eval_loss": 0.7630029916763306,
|
| 599 |
+
"eval_runtime": 1.1689,
|
| 600 |
+
"eval_samples_per_second": 97.529,
|
| 601 |
+
"eval_steps_per_second": 5.133,
|
| 602 |
+
"step": 390
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.2307692307692308,
|
| 606 |
+
"grad_norm": 2.3290536403656006,
|
| 607 |
+
"learning_rate": 1.9967573081342103e-05,
|
| 608 |
+
"loss": 0.4174,
|
| 609 |
+
"step": 400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.2307692307692308,
|
| 613 |
+
"eval_loss": 0.7669293880462646,
|
| 614 |
+
"eval_runtime": 1.1711,
|
| 615 |
+
"eval_samples_per_second": 97.345,
|
| 616 |
+
"eval_steps_per_second": 5.123,
|
| 617 |
+
"step": 400
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.2615384615384615,
|
| 621 |
+
"grad_norm": 2.4483697414398193,
|
| 622 |
+
"learning_rate": 1.9958355831085155e-05,
|
| 623 |
+
"loss": 0.4045,
|
| 624 |
+
"step": 410
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.2615384615384615,
|
| 628 |
+
"eval_loss": 0.7652024626731873,
|
| 629 |
+
"eval_runtime": 1.1697,
|
| 630 |
+
"eval_samples_per_second": 97.461,
|
| 631 |
+
"eval_steps_per_second": 5.13,
|
| 632 |
+
"step": 410
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.2923076923076924,
|
| 636 |
+
"grad_norm": 1.703616976737976,
|
| 637 |
+
"learning_rate": 1.9947989815101444e-05,
|
| 638 |
+
"loss": 0.4292,
|
| 639 |
+
"step": 420
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.2923076923076924,
|
| 643 |
+
"eval_loss": 0.7553891539573669,
|
| 644 |
+
"eval_runtime": 1.1719,
|
| 645 |
+
"eval_samples_per_second": 97.277,
|
| 646 |
+
"eval_steps_per_second": 5.12,
|
| 647 |
+
"step": 420
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.323076923076923,
|
| 651 |
+
"grad_norm": 1.6245399713516235,
|
| 652 |
+
"learning_rate": 1.9936476229183133e-05,
|
| 653 |
+
"loss": 0.4164,
|
| 654 |
+
"step": 430
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.323076923076923,
|
| 658 |
+
"eval_loss": 0.7711736559867859,
|
| 659 |
+
"eval_runtime": 1.1776,
|
| 660 |
+
"eval_samples_per_second": 96.806,
|
| 661 |
+
"eval_steps_per_second": 5.095,
|
| 662 |
+
"step": 430
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.353846153846154,
|
| 666 |
+
"grad_norm": 1.9511611461639404,
|
| 667 |
+
"learning_rate": 1.992381640150257e-05,
|
| 668 |
+
"loss": 0.4337,
|
| 669 |
+
"step": 440
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.353846153846154,
|
| 673 |
+
"eval_loss": 0.7438372373580933,
|
| 674 |
+
"eval_runtime": 1.1692,
|
| 675 |
+
"eval_samples_per_second": 97.5,
|
| 676 |
+
"eval_steps_per_second": 5.132,
|
| 677 |
+
"step": 440
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.3846153846153846,
|
| 681 |
+
"grad_norm": 2.466737747192383,
|
| 682 |
+
"learning_rate": 1.9910011792459086e-05,
|
| 683 |
+
"loss": 0.3712,
|
| 684 |
+
"step": 450
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.3846153846153846,
|
| 688 |
+
"eval_loss": 0.8045614361763,
|
| 689 |
+
"eval_runtime": 1.1689,
|
| 690 |
+
"eval_samples_per_second": 97.53,
|
| 691 |
+
"eval_steps_per_second": 5.133,
|
| 692 |
+
"step": 450
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.4153846153846155,
|
| 696 |
+
"grad_norm": 2.473851203918457,
|
| 697 |
+
"learning_rate": 1.9895063994510512e-05,
|
| 698 |
+
"loss": 0.516,
|
| 699 |
+
"step": 460
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.4153846153846155,
|
| 703 |
+
"eval_loss": 0.7463352084159851,
|
| 704 |
+
"eval_runtime": 1.1697,
|
| 705 |
+
"eval_samples_per_second": 97.464,
|
| 706 |
+
"eval_steps_per_second": 5.13,
|
| 707 |
+
"step": 460
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.4461538461538461,
|
| 711 |
+
"grad_norm": 2.757955312728882,
|
| 712 |
+
"learning_rate": 1.9878974731989487e-05,
|
| 713 |
+
"loss": 0.4965,
|
| 714 |
+
"step": 470
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.4461538461538461,
|
| 718 |
+
"eval_loss": 0.7571713924407959,
|
| 719 |
+
"eval_runtime": 1.1688,
|
| 720 |
+
"eval_samples_per_second": 97.534,
|
| 721 |
+
"eval_steps_per_second": 5.133,
|
| 722 |
+
"step": 470
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.476923076923077,
|
| 726 |
+
"grad_norm": 3.9884915351867676,
|
| 727 |
+
"learning_rate": 1.9861745860904538e-05,
|
| 728 |
+
"loss": 0.3973,
|
| 729 |
+
"step": 480
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.476923076923077,
|
| 733 |
+
"eval_loss": 0.7763081789016724,
|
| 734 |
+
"eval_runtime": 1.168,
|
| 735 |
+
"eval_samples_per_second": 97.605,
|
| 736 |
+
"eval_steps_per_second": 5.137,
|
| 737 |
+
"step": 480
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5076923076923077,
|
| 741 |
+
"grad_norm": 1.9474631547927856,
|
| 742 |
+
"learning_rate": 1.9843379368725978e-05,
|
| 743 |
+
"loss": 0.4625,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5076923076923077,
|
| 748 |
+
"eval_loss": 0.7595367431640625,
|
| 749 |
+
"eval_runtime": 1.1708,
|
| 750 |
+
"eval_samples_per_second": 97.37,
|
| 751 |
+
"eval_steps_per_second": 5.125,
|
| 752 |
+
"step": 490
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5384615384615383,
|
| 756 |
+
"grad_norm": 2.2907869815826416,
|
| 757 |
+
"learning_rate": 1.9823877374156647e-05,
|
| 758 |
+
"loss": 0.4371,
|
| 759 |
+
"step": 500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5384615384615383,
|
| 763 |
+
"eval_loss": 0.762865424156189,
|
| 764 |
+
"eval_runtime": 1.1766,
|
| 765 |
+
"eval_samples_per_second": 96.891,
|
| 766 |
+
"eval_steps_per_second": 5.1,
|
| 767 |
+
"step": 500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.5692307692307692,
|
| 771 |
+
"grad_norm": 3.067608594894409,
|
| 772 |
+
"learning_rate": 1.9803242126887496e-05,
|
| 773 |
+
"loss": 0.45,
|
| 774 |
+
"step": 510
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.5692307692307692,
|
| 778 |
+
"eval_loss": 0.7638718485832214,
|
| 779 |
+
"eval_runtime": 1.1684,
|
| 780 |
+
"eval_samples_per_second": 97.568,
|
| 781 |
+
"eval_steps_per_second": 5.135,
|
| 782 |
+
"step": 510
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.6,
|
| 786 |
+
"grad_norm": 2.1113197803497314,
|
| 787 |
+
"learning_rate": 1.9781476007338058e-05,
|
| 788 |
+
"loss": 0.4524,
|
| 789 |
+
"step": 520
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.6,
|
| 793 |
+
"eval_loss": 0.7560412883758545,
|
| 794 |
+
"eval_runtime": 1.1719,
|
| 795 |
+
"eval_samples_per_second": 97.278,
|
| 796 |
+
"eval_steps_per_second": 5.12,
|
| 797 |
+
"step": 520
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.6307692307692307,
|
| 801 |
+
"grad_norm": 1.889585256576538,
|
| 802 |
+
"learning_rate": 1.9758581526381878e-05,
|
| 803 |
+
"loss": 0.4287,
|
| 804 |
+
"step": 530
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.6307692307692307,
|
| 808 |
+
"eval_loss": 0.7591350674629211,
|
| 809 |
+
"eval_runtime": 1.1693,
|
| 810 |
+
"eval_samples_per_second": 97.495,
|
| 811 |
+
"eval_steps_per_second": 5.131,
|
| 812 |
+
"step": 530
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.6615384615384614,
|
| 816 |
+
"grad_norm": 2.519418716430664,
|
| 817 |
+
"learning_rate": 1.973456132505684e-05,
|
| 818 |
+
"loss": 0.4603,
|
| 819 |
+
"step": 540
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.6615384615384614,
|
| 823 |
+
"eval_loss": 0.7536805868148804,
|
| 824 |
+
"eval_runtime": 1.1799,
|
| 825 |
+
"eval_samples_per_second": 96.617,
|
| 826 |
+
"eval_steps_per_second": 5.085,
|
| 827 |
+
"step": 540
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.6923076923076923,
|
| 831 |
+
"grad_norm": 2.7455596923828125,
|
| 832 |
+
"learning_rate": 1.9709418174260523e-05,
|
| 833 |
+
"loss": 0.4293,
|
| 834 |
+
"step": 550
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.6923076923076923,
|
| 838 |
+
"eval_loss": 0.763169527053833,
|
| 839 |
+
"eval_runtime": 1.1696,
|
| 840 |
+
"eval_samples_per_second": 97.47,
|
| 841 |
+
"eval_steps_per_second": 5.13,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7230769230769232,
|
| 846 |
+
"grad_norm": 1.9823411703109741,
|
| 847 |
+
"learning_rate": 1.9683154974430544e-05,
|
| 848 |
+
"loss": 0.4643,
|
| 849 |
+
"step": 560
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7230769230769232,
|
| 853 |
+
"eval_loss": 0.7607314586639404,
|
| 854 |
+
"eval_runtime": 1.1712,
|
| 855 |
+
"eval_samples_per_second": 97.336,
|
| 856 |
+
"eval_steps_per_second": 5.123,
|
| 857 |
+
"step": 560
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.7538461538461538,
|
| 861 |
+
"grad_norm": 2.2437684535980225,
|
| 862 |
+
"learning_rate": 1.965577475520999e-05,
|
| 863 |
+
"loss": 0.452,
|
| 864 |
+
"step": 570
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.7538461538461538,
|
| 868 |
+
"eval_loss": 0.7615717649459839,
|
| 869 |
+
"eval_runtime": 1.1946,
|
| 870 |
+
"eval_samples_per_second": 95.425,
|
| 871 |
+
"eval_steps_per_second": 5.022,
|
| 872 |
+
"step": 570
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.7846153846153845,
|
| 876 |
+
"grad_norm": 3.0740182399749756,
|
| 877 |
+
"learning_rate": 1.962728067509791e-05,
|
| 878 |
+
"loss": 0.4885,
|
| 879 |
+
"step": 580
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.7846153846153845,
|
| 883 |
+
"eval_loss": 0.7764689326286316,
|
| 884 |
+
"eval_runtime": 1.1714,
|
| 885 |
+
"eval_samples_per_second": 97.32,
|
| 886 |
+
"eval_steps_per_second": 5.122,
|
| 887 |
+
"step": 580
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.8153846153846154,
|
| 891 |
+
"grad_norm": 2.050313711166382,
|
| 892 |
+
"learning_rate": 1.9597676021084962e-05,
|
| 893 |
+
"loss": 0.4992,
|
| 894 |
+
"step": 590
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.8153846153846154,
|
| 898 |
+
"eval_loss": 0.7457339763641357,
|
| 899 |
+
"eval_runtime": 1.1747,
|
| 900 |
+
"eval_samples_per_second": 97.045,
|
| 901 |
+
"eval_steps_per_second": 5.108,
|
| 902 |
+
"step": 590
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.8461538461538463,
|
| 906 |
+
"grad_norm": 2.349470853805542,
|
| 907 |
+
"learning_rate": 1.9566964208274254e-05,
|
| 908 |
+
"loss": 0.3963,
|
| 909 |
+
"step": 600
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.8461538461538463,
|
| 913 |
+
"eval_loss": 0.7712545990943909,
|
| 914 |
+
"eval_runtime": 1.17,
|
| 915 |
+
"eval_samples_per_second": 97.433,
|
| 916 |
+
"eval_steps_per_second": 5.128,
|
| 917 |
+
"step": 600
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.876923076923077,
|
| 921 |
+
"grad_norm": 1.947912573814392,
|
| 922 |
+
"learning_rate": 1.9535148779487365e-05,
|
| 923 |
+
"loss": 0.4388,
|
| 924 |
+
"step": 610
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.876923076923077,
|
| 928 |
+
"eval_loss": 0.772519052028656,
|
| 929 |
+
"eval_runtime": 1.1717,
|
| 930 |
+
"eval_samples_per_second": 97.296,
|
| 931 |
+
"eval_steps_per_second": 5.121,
|
| 932 |
+
"step": 610
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.9076923076923076,
|
| 936 |
+
"grad_norm": 1.8837225437164307,
|
| 937 |
+
"learning_rate": 1.9502233404855672e-05,
|
| 938 |
+
"loss": 0.3965,
|
| 939 |
+
"step": 620
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.9076923076923076,
|
| 943 |
+
"eval_loss": 0.7594517469406128,
|
| 944 |
+
"eval_runtime": 1.2215,
|
| 945 |
+
"eval_samples_per_second": 93.326,
|
| 946 |
+
"eval_steps_per_second": 4.912,
|
| 947 |
+
"step": 620
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9384615384615385,
|
| 951 |
+
"grad_norm": 1.8398792743682861,
|
| 952 |
+
"learning_rate": 1.946822188139696e-05,
|
| 953 |
+
"loss": 0.3898,
|
| 954 |
+
"step": 630
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.9384615384615385,
|
| 958 |
+
"eval_loss": 0.7748451828956604,
|
| 959 |
+
"eval_runtime": 1.2222,
|
| 960 |
+
"eval_samples_per_second": 93.273,
|
| 961 |
+
"eval_steps_per_second": 4.909,
|
| 962 |
+
"step": 630
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.9692307692307693,
|
| 966 |
+
"grad_norm": 2.3971219062805176,
|
| 967 |
+
"learning_rate": 1.9433118132577432e-05,
|
| 968 |
+
"loss": 0.4253,
|
| 969 |
+
"step": 640
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.9692307692307693,
|
| 973 |
+
"eval_loss": 0.7729252576828003,
|
| 974 |
+
"eval_runtime": 1.1743,
|
| 975 |
+
"eval_samples_per_second": 97.081,
|
| 976 |
+
"eval_steps_per_second": 5.11,
|
| 977 |
+
"step": 640
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.0,
|
| 981 |
+
"grad_norm": 1.657328724861145,
|
| 982 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 983 |
+
"loss": 0.4898,
|
| 984 |
+
"step": 650
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.0,
|
| 988 |
+
"eval_loss": 0.7602188587188721,
|
| 989 |
+
"eval_runtime": 1.1738,
|
| 990 |
+
"eval_samples_per_second": 97.117,
|
| 991 |
+
"eval_steps_per_second": 5.111,
|
| 992 |
+
"step": 650
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.0307692307692307,
|
| 996 |
+
"grad_norm": 1.3759773969650269,
|
| 997 |
+
"learning_rate": 1.935965028223259e-05,
|
| 998 |
+
"loss": 0.2182,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.0307692307692307,
|
| 1003 |
+
"eval_loss": 0.8499692678451538,
|
| 1004 |
+
"eval_runtime": 1.1719,
|
| 1005 |
+
"eval_samples_per_second": 97.276,
|
| 1006 |
+
"eval_steps_per_second": 5.12,
|
| 1007 |
+
"step": 660
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.0615384615384613,
|
| 1011 |
+
"grad_norm": 1.4757074117660522,
|
| 1012 |
+
"learning_rate": 1.932129465573568e-05,
|
| 1013 |
+
"loss": 0.2196,
|
| 1014 |
+
"step": 670
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.0615384615384613,
|
| 1018 |
+
"eval_loss": 0.866599977016449,
|
| 1019 |
+
"eval_runtime": 1.1739,
|
| 1020 |
+
"eval_samples_per_second": 97.115,
|
| 1021 |
+
"eval_steps_per_second": 5.111,
|
| 1022 |
+
"step": 670
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.0923076923076924,
|
| 1026 |
+
"grad_norm": 1.2500200271606445,
|
| 1027 |
+
"learning_rate": 1.9281863752957095e-05,
|
| 1028 |
+
"loss": 0.1772,
|
| 1029 |
+
"step": 680
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.0923076923076924,
|
| 1033 |
+
"eval_loss": 0.8642630577087402,
|
| 1034 |
+
"eval_runtime": 1.1753,
|
| 1035 |
+
"eval_samples_per_second": 96.999,
|
| 1036 |
+
"eval_steps_per_second": 5.105,
|
| 1037 |
+
"step": 680
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.123076923076923,
|
| 1041 |
+
"grad_norm": 2.561828851699829,
|
| 1042 |
+
"learning_rate": 1.92413621225262e-05,
|
| 1043 |
+
"loss": 0.2014,
|
| 1044 |
+
"step": 690
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.123076923076923,
|
| 1048 |
+
"eval_loss": 0.8872668743133545,
|
| 1049 |
+
"eval_runtime": 1.1761,
|
| 1050 |
+
"eval_samples_per_second": 96.933,
|
| 1051 |
+
"eval_steps_per_second": 5.102,
|
| 1052 |
+
"step": 690
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.1538461538461537,
|
| 1056 |
+
"grad_norm": 1.8431521654129028,
|
| 1057 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 1058 |
+
"loss": 0.2245,
|
| 1059 |
+
"step": 700
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.1538461538461537,
|
| 1063 |
+
"eval_loss": 0.8799911737442017,
|
| 1064 |
+
"eval_runtime": 1.1766,
|
| 1065 |
+
"eval_samples_per_second": 96.89,
|
| 1066 |
+
"eval_steps_per_second": 5.099,
|
| 1067 |
+
"step": 700
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.184615384615385,
|
| 1071 |
+
"grad_norm": 1.4718531370162964,
|
| 1072 |
+
"learning_rate": 1.915716549026541e-05,
|
| 1073 |
+
"loss": 0.1911,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.184615384615385,
|
| 1078 |
+
"eval_loss": 0.8806982636451721,
|
| 1079 |
+
"eval_runtime": 1.1696,
|
| 1080 |
+
"eval_samples_per_second": 97.473,
|
| 1081 |
+
"eval_steps_per_second": 5.13,
|
| 1082 |
+
"step": 710
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.2153846153846155,
|
| 1086 |
+
"grad_norm": 2.061520576477051,
|
| 1087 |
+
"learning_rate": 1.9113480201103658e-05,
|
| 1088 |
+
"loss": 0.232,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.2153846153846155,
|
| 1093 |
+
"eval_loss": 0.8812103271484375,
|
| 1094 |
+
"eval_runtime": 1.1676,
|
| 1095 |
+
"eval_samples_per_second": 97.636,
|
| 1096 |
+
"eval_steps_per_second": 5.139,
|
| 1097 |
+
"step": 720
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.246153846153846,
|
| 1101 |
+
"grad_norm": 2.530285596847534,
|
| 1102 |
+
"learning_rate": 1.9068743608505454e-05,
|
| 1103 |
+
"loss": 0.2232,
|
| 1104 |
+
"step": 730
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.246153846153846,
|
| 1108 |
+
"eval_loss": 0.8682798147201538,
|
| 1109 |
+
"eval_runtime": 1.1687,
|
| 1110 |
+
"eval_samples_per_second": 97.545,
|
| 1111 |
+
"eval_steps_per_second": 5.134,
|
| 1112 |
+
"step": 730
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.276923076923077,
|
| 1116 |
+
"grad_norm": 1.8433207273483276,
|
| 1117 |
+
"learning_rate": 1.902296087314845e-05,
|
| 1118 |
+
"loss": 0.2315,
|
| 1119 |
+
"step": 740
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.276923076923077,
|
| 1123 |
+
"eval_loss": 0.8747490048408508,
|
| 1124 |
+
"eval_runtime": 1.1702,
|
| 1125 |
+
"eval_samples_per_second": 97.421,
|
| 1126 |
+
"eval_steps_per_second": 5.127,
|
| 1127 |
+
"step": 740
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.3076923076923075,
|
| 1131 |
+
"grad_norm": 2.1529664993286133,
|
| 1132 |
+
"learning_rate": 1.8976137276390145e-05,
|
| 1133 |
+
"loss": 0.239,
|
| 1134 |
+
"step": 750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.3076923076923075,
|
| 1138 |
+
"eval_loss": 0.8778530359268188,
|
| 1139 |
+
"eval_runtime": 1.1674,
|
| 1140 |
+
"eval_samples_per_second": 97.656,
|
| 1141 |
+
"eval_steps_per_second": 5.14,
|
| 1142 |
+
"step": 750
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.3384615384615386,
|
| 1146 |
+
"grad_norm": 1.839979887008667,
|
| 1147 |
+
"learning_rate": 1.892827821965864e-05,
|
| 1148 |
+
"loss": 0.2228,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.3384615384615386,
|
| 1153 |
+
"eval_loss": 0.8769873976707458,
|
| 1154 |
+
"eval_runtime": 1.1718,
|
| 1155 |
+
"eval_samples_per_second": 97.288,
|
| 1156 |
+
"eval_steps_per_second": 5.12,
|
| 1157 |
+
"step": 760
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.3692307692307693,
|
| 1161 |
+
"grad_norm": 1.3814674615859985,
|
| 1162 |
+
"learning_rate": 1.8879389223829592e-05,
|
| 1163 |
+
"loss": 0.1999,
|
| 1164 |
+
"step": 770
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.3692307692307693,
|
| 1168 |
+
"eval_loss": 0.8731983304023743,
|
| 1169 |
+
"eval_runtime": 1.1723,
|
| 1170 |
+
"eval_samples_per_second": 97.242,
|
| 1171 |
+
"eval_steps_per_second": 5.118,
|
| 1172 |
+
"step": 770
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.4,
|
| 1176 |
+
"grad_norm": 1.5382564067840576,
|
| 1177 |
+
"learning_rate": 1.8829475928589272e-05,
|
| 1178 |
+
"loss": 0.1888,
|
| 1179 |
+
"step": 780
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.4,
|
| 1183 |
+
"eval_loss": 0.8794592022895813,
|
| 1184 |
+
"eval_runtime": 1.1678,
|
| 1185 |
+
"eval_samples_per_second": 97.622,
|
| 1186 |
+
"eval_steps_per_second": 5.138,
|
| 1187 |
+
"step": 780
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.430769230769231,
|
| 1191 |
+
"grad_norm": 2.371277093887329,
|
| 1192 |
+
"learning_rate": 1.8778544091784047e-05,
|
| 1193 |
+
"loss": 0.2293,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.430769230769231,
|
| 1198 |
+
"eval_loss": 0.8893064260482788,
|
| 1199 |
+
"eval_runtime": 1.1684,
|
| 1200 |
+
"eval_samples_per_second": 97.567,
|
| 1201 |
+
"eval_steps_per_second": 5.135,
|
| 1202 |
+
"step": 790
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.4615384615384617,
|
| 1206 |
+
"grad_norm": 1.4892629384994507,
|
| 1207 |
+
"learning_rate": 1.8726599588756144e-05,
|
| 1208 |
+
"loss": 0.2326,
|
| 1209 |
+
"step": 800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.4615384615384617,
|
| 1213 |
+
"eval_loss": 0.8676194548606873,
|
| 1214 |
+
"eval_runtime": 1.1699,
|
| 1215 |
+
"eval_samples_per_second": 97.441,
|
| 1216 |
+
"eval_steps_per_second": 5.128,
|
| 1217 |
+
"step": 800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.4923076923076923,
|
| 1221 |
+
"grad_norm": 1.6047152280807495,
|
| 1222 |
+
"learning_rate": 1.8673648411665895e-05,
|
| 1223 |
+
"loss": 0.1986,
|
| 1224 |
+
"step": 810
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.4923076923076923,
|
| 1228 |
+
"eval_loss": 0.8942193984985352,
|
| 1229 |
+
"eval_runtime": 1.17,
|
| 1230 |
+
"eval_samples_per_second": 97.433,
|
| 1231 |
+
"eval_steps_per_second": 5.128,
|
| 1232 |
+
"step": 810
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.523076923076923,
|
| 1236 |
+
"grad_norm": 2.106461524963379,
|
| 1237 |
+
"learning_rate": 1.8619696668800494e-05,
|
| 1238 |
+
"loss": 0.238,
|
| 1239 |
+
"step": 820
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.523076923076923,
|
| 1243 |
+
"eval_loss": 0.8907297849655151,
|
| 1244 |
+
"eval_runtime": 1.1681,
|
| 1245 |
+
"eval_samples_per_second": 97.593,
|
| 1246 |
+
"eval_steps_per_second": 5.136,
|
| 1247 |
+
"step": 820
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 2.5538461538461537,
|
| 1251 |
+
"grad_norm": 1.7706941366195679,
|
| 1252 |
+
"learning_rate": 1.8564750583869374e-05,
|
| 1253 |
+
"loss": 0.2278,
|
| 1254 |
+
"step": 830
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 2.5538461538461537,
|
| 1258 |
+
"eval_loss": 0.8588199019432068,
|
| 1259 |
+
"eval_runtime": 1.1709,
|
| 1260 |
+
"eval_samples_per_second": 97.36,
|
| 1261 |
+
"eval_steps_per_second": 5.124,
|
| 1262 |
+
"step": 830
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.5846153846153848,
|
| 1266 |
+
"grad_norm": 1.4446183443069458,
|
| 1267 |
+
"learning_rate": 1.850881649528625e-05,
|
| 1268 |
+
"loss": 0.2033,
|
| 1269 |
+
"step": 840
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.5846153846153848,
|
| 1273 |
+
"eval_loss": 0.8720155954360962,
|
| 1274 |
+
"eval_runtime": 1.1716,
|
| 1275 |
+
"eval_samples_per_second": 97.304,
|
| 1276 |
+
"eval_steps_per_second": 5.121,
|
| 1277 |
+
"step": 840
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.6153846153846154,
|
| 1281 |
+
"grad_norm": 1.5155885219573975,
|
| 1282 |
+
"learning_rate": 1.845190085543795e-05,
|
| 1283 |
+
"loss": 0.2211,
|
| 1284 |
+
"step": 850
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6153846153846154,
|
| 1288 |
+
"eval_loss": 0.8732415437698364,
|
| 1289 |
+
"eval_runtime": 1.1708,
|
| 1290 |
+
"eval_samples_per_second": 97.371,
|
| 1291 |
+
"eval_steps_per_second": 5.125,
|
| 1292 |
+
"step": 850
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.646153846153846,
|
| 1296 |
+
"grad_norm": 2.934957981109619,
|
| 1297 |
+
"learning_rate": 1.839401022994006e-05,
|
| 1298 |
+
"loss": 0.21,
|
| 1299 |
+
"step": 860
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.646153846153846,
|
| 1303 |
+
"eval_loss": 0.8790932297706604,
|
| 1304 |
+
"eval_runtime": 1.1671,
|
| 1305 |
+
"eval_samples_per_second": 97.68,
|
| 1306 |
+
"eval_steps_per_second": 5.141,
|
| 1307 |
+
"step": 860
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.676923076923077,
|
| 1311 |
+
"grad_norm": 2.930457353591919,
|
| 1312 |
+
"learning_rate": 1.8335151296879576e-05,
|
| 1313 |
+
"loss": 0.2313,
|
| 1314 |
+
"step": 870
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.676923076923077,
|
| 1318 |
+
"eval_loss": 0.8555310368537903,
|
| 1319 |
+
"eval_runtime": 1.1718,
|
| 1320 |
+
"eval_samples_per_second": 97.284,
|
| 1321 |
+
"eval_steps_per_second": 5.12,
|
| 1322 |
+
"step": 870
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.707692307692308,
|
| 1326 |
+
"grad_norm": 2.029303789138794,
|
| 1327 |
+
"learning_rate": 1.82753308460445e-05,
|
| 1328 |
+
"loss": 0.2285,
|
| 1329 |
+
"step": 880
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.707692307692308,
|
| 1333 |
+
"eval_loss": 0.870873749256134,
|
| 1334 |
+
"eval_runtime": 1.1696,
|
| 1335 |
+
"eval_samples_per_second": 97.47,
|
| 1336 |
+
"eval_steps_per_second": 5.13,
|
| 1337 |
+
"step": 880
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 2.7384615384615385,
|
| 1341 |
+
"grad_norm": 1.949743390083313,
|
| 1342 |
+
"learning_rate": 1.821455577814062e-05,
|
| 1343 |
+
"loss": 0.2134,
|
| 1344 |
+
"step": 890
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.7384615384615385,
|
| 1348 |
+
"eval_loss": 0.8873838186264038,
|
| 1349 |
+
"eval_runtime": 1.1688,
|
| 1350 |
+
"eval_samples_per_second": 97.538,
|
| 1351 |
+
"eval_steps_per_second": 5.134,
|
| 1352 |
+
"step": 890
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.769230769230769,
|
| 1356 |
+
"grad_norm": 1.6945637464523315,
|
| 1357 |
+
"learning_rate": 1.8152833103995443e-05,
|
| 1358 |
+
"loss": 0.218,
|
| 1359 |
+
"step": 900
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.769230769230769,
|
| 1363 |
+
"eval_loss": 0.8798539042472839,
|
| 1364 |
+
"eval_runtime": 1.1699,
|
| 1365 |
+
"eval_samples_per_second": 97.444,
|
| 1366 |
+
"eval_steps_per_second": 5.129,
|
| 1367 |
+
"step": 900
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8,
|
| 1371 |
+
"grad_norm": 2.2784035205841064,
|
| 1372 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 1373 |
+
"loss": 0.2426,
|
| 1374 |
+
"step": 910
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8,
|
| 1378 |
+
"eval_loss": 0.8753871321678162,
|
| 1379 |
+
"eval_runtime": 1.1723,
|
| 1380 |
+
"eval_samples_per_second": 97.246,
|
| 1381 |
+
"eval_steps_per_second": 5.118,
|
| 1382 |
+
"step": 910
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.830769230769231,
|
| 1386 |
+
"grad_norm": 1.750718593597412,
|
| 1387 |
+
"learning_rate": 1.802657352603483e-05,
|
| 1388 |
+
"loss": 0.2059,
|
| 1389 |
+
"step": 920
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.830769230769231,
|
| 1393 |
+
"eval_loss": 0.8761821985244751,
|
| 1394 |
+
"eval_runtime": 1.1762,
|
| 1395 |
+
"eval_samples_per_second": 96.92,
|
| 1396 |
+
"eval_steps_per_second": 5.101,
|
| 1397 |
+
"step": 920
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.8615384615384616,
|
| 1401 |
+
"grad_norm": 1.2204371690750122,
|
| 1402 |
+
"learning_rate": 1.7962051187141377e-05,
|
| 1403 |
+
"loss": 0.2229,
|
| 1404 |
+
"step": 930
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.8615384615384616,
|
| 1408 |
+
"eval_loss": 0.8772630095481873,
|
| 1409 |
+
"eval_runtime": 1.1697,
|
| 1410 |
+
"eval_samples_per_second": 97.458,
|
| 1411 |
+
"eval_steps_per_second": 5.129,
|
| 1412 |
+
"step": 930
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.8923076923076922,
|
| 1416 |
+
"grad_norm": 2.0748281478881836,
|
| 1417 |
+
"learning_rate": 1.7896610370170452e-05,
|
| 1418 |
+
"loss": 0.2582,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.8923076923076922,
|
| 1423 |
+
"eval_loss": 0.884286105632782,
|
| 1424 |
+
"eval_runtime": 1.1773,
|
| 1425 |
+
"eval_samples_per_second": 96.829,
|
| 1426 |
+
"eval_steps_per_second": 5.096,
|
| 1427 |
+
"step": 940
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.9230769230769234,
|
| 1431 |
+
"grad_norm": 1.9051027297973633,
|
| 1432 |
+
"learning_rate": 1.7830258624176224e-05,
|
| 1433 |
+
"loss": 0.2102,
|
| 1434 |
+
"step": 950
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.9230769230769234,
|
| 1438 |
+
"eval_loss": 0.878767192363739,
|
| 1439 |
+
"eval_runtime": 1.1684,
|
| 1440 |
+
"eval_samples_per_second": 97.571,
|
| 1441 |
+
"eval_steps_per_second": 5.135,
|
| 1442 |
+
"step": 950
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.953846153846154,
|
| 1446 |
+
"grad_norm": 1.8921849727630615,
|
| 1447 |
+
"learning_rate": 1.776300360329488e-05,
|
| 1448 |
+
"loss": 0.2331,
|
| 1449 |
+
"step": 960
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.953846153846154,
|
| 1453 |
+
"eval_loss": 0.8681402206420898,
|
| 1454 |
+
"eval_runtime": 1.1693,
|
| 1455 |
+
"eval_samples_per_second": 97.497,
|
| 1456 |
+
"eval_steps_per_second": 5.131,
|
| 1457 |
+
"step": 960
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.9846153846153847,
|
| 1461 |
+
"grad_norm": 1.5593761205673218,
|
| 1462 |
+
"learning_rate": 1.769485306586166e-05,
|
| 1463 |
+
"loss": 0.2226,
|
| 1464 |
+
"step": 970
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.9846153846153847,
|
| 1468 |
+
"eval_loss": 0.8726215958595276,
|
| 1469 |
+
"eval_runtime": 1.1712,
|
| 1470 |
+
"eval_samples_per_second": 97.336,
|
| 1471 |
+
"eval_steps_per_second": 5.123,
|
| 1472 |
+
"step": 970
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.0153846153846153,
|
| 1476 |
+
"grad_norm": 1.4075913429260254,
|
| 1477 |
+
"learning_rate": 1.762581487351587e-05,
|
| 1478 |
+
"loss": 0.1535,
|
| 1479 |
+
"step": 980
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.0153846153846153,
|
| 1483 |
+
"eval_loss": 0.9113463163375854,
|
| 1484 |
+
"eval_runtime": 1.1794,
|
| 1485 |
+
"eval_samples_per_second": 96.661,
|
| 1486 |
+
"eval_steps_per_second": 5.087,
|
| 1487 |
+
"step": 980
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.046153846153846,
|
| 1491 |
+
"grad_norm": 1.410906434059143,
|
| 1492 |
+
"learning_rate": 1.7555896990294003e-05,
|
| 1493 |
+
"loss": 0.1021,
|
| 1494 |
+
"step": 990
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.046153846153846,
|
| 1498 |
+
"eval_loss": 0.9633604288101196,
|
| 1499 |
+
"eval_runtime": 1.1702,
|
| 1500 |
+
"eval_samples_per_second": 97.416,
|
| 1501 |
+
"eval_steps_per_second": 5.127,
|
| 1502 |
+
"step": 990
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 3.076923076923077,
|
| 1506 |
+
"grad_norm": 1.0470328330993652,
|
| 1507 |
+
"learning_rate": 1.7485107481711014e-05,
|
| 1508 |
+
"loss": 0.1118,
|
| 1509 |
+
"step": 1000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 3.076923076923077,
|
| 1513 |
+
"eval_loss": 0.9647616744041443,
|
| 1514 |
+
"eval_runtime": 1.1683,
|
| 1515 |
+
"eval_samples_per_second": 97.574,
|
| 1516 |
+
"eval_steps_per_second": 5.135,
|
| 1517 |
+
"step": 1000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.1076923076923078,
|
| 1521 |
+
"grad_norm": 2.098592758178711,
|
| 1522 |
+
"learning_rate": 1.741345451382992e-05,
|
| 1523 |
+
"loss": 0.1224,
|
| 1524 |
+
"step": 1010
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.1076923076923078,
|
| 1528 |
+
"eval_loss": 0.9478728175163269,
|
| 1529 |
+
"eval_runtime": 1.1735,
|
| 1530 |
+
"eval_samples_per_second": 97.143,
|
| 1531 |
+
"eval_steps_per_second": 5.113,
|
| 1532 |
+
"step": 1010
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 3.1384615384615384,
|
| 1536 |
+
"grad_norm": 1.6263700723648071,
|
| 1537 |
+
"learning_rate": 1.7340946352319795e-05,
|
| 1538 |
+
"loss": 0.1377,
|
| 1539 |
+
"step": 1020
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 3.1384615384615384,
|
| 1543 |
+
"eval_loss": 0.922594428062439,
|
| 1544 |
+
"eval_runtime": 1.1721,
|
| 1545 |
+
"eval_samples_per_second": 97.262,
|
| 1546 |
+
"eval_steps_per_second": 5.119,
|
| 1547 |
+
"step": 1020
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 3.169230769230769,
|
| 1551 |
+
"grad_norm": 1.0413360595703125,
|
| 1552 |
+
"learning_rate": 1.7267591361502233e-05,
|
| 1553 |
+
"loss": 0.1249,
|
| 1554 |
+
"step": 1030
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 3.169230769230769,
|
| 1558 |
+
"eval_loss": 0.9370183944702148,
|
| 1559 |
+
"eval_runtime": 1.169,
|
| 1560 |
+
"eval_samples_per_second": 97.519,
|
| 1561 |
+
"eval_steps_per_second": 5.133,
|
| 1562 |
+
"step": 1030
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 3.2,
|
| 1566 |
+
"grad_norm": 1.6745877265930176,
|
| 1567 |
+
"learning_rate": 1.7193398003386514e-05,
|
| 1568 |
+
"loss": 0.1186,
|
| 1569 |
+
"step": 1040
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 3.2,
|
| 1573 |
+
"eval_loss": 0.9573487639427185,
|
| 1574 |
+
"eval_runtime": 1.1681,
|
| 1575 |
+
"eval_samples_per_second": 97.596,
|
| 1576 |
+
"eval_steps_per_second": 5.137,
|
| 1577 |
+
"step": 1040
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.230769230769231,
|
| 1581 |
+
"grad_norm": 1.3979861736297607,
|
| 1582 |
+
"learning_rate": 1.7118374836693407e-05,
|
| 1583 |
+
"loss": 0.135,
|
| 1584 |
+
"step": 1050
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.230769230769231,
|
| 1588 |
+
"eval_loss": 0.9554935693740845,
|
| 1589 |
+
"eval_runtime": 1.1664,
|
| 1590 |
+
"eval_samples_per_second": 97.735,
|
| 1591 |
+
"eval_steps_per_second": 5.144,
|
| 1592 |
+
"step": 1050
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 3.2615384615384615,
|
| 1596 |
+
"grad_norm": 1.4017126560211182,
|
| 1597 |
+
"learning_rate": 1.7042530515867897e-05,
|
| 1598 |
+
"loss": 0.1102,
|
| 1599 |
+
"step": 1060
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 3.2615384615384615,
|
| 1603 |
+
"eval_loss": 0.9547311663627625,
|
| 1604 |
+
"eval_runtime": 1.1685,
|
| 1605 |
+
"eval_samples_per_second": 97.561,
|
| 1606 |
+
"eval_steps_per_second": 5.135,
|
| 1607 |
+
"step": 1060
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 3.292307692307692,
|
| 1611 |
+
"grad_norm": 1.3390568494796753,
|
| 1612 |
+
"learning_rate": 1.6965873790080806e-05,
|
| 1613 |
+
"loss": 0.1358,
|
| 1614 |
+
"step": 1070
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 3.292307692307692,
|
| 1618 |
+
"eval_loss": 0.9695214629173279,
|
| 1619 |
+
"eval_runtime": 1.1695,
|
| 1620 |
+
"eval_samples_per_second": 97.474,
|
| 1621 |
+
"eval_steps_per_second": 5.13,
|
| 1622 |
+
"step": 1070
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 3.3230769230769233,
|
| 1626 |
+
"grad_norm": 1.9002848863601685,
|
| 1627 |
+
"learning_rate": 1.6888413502219534e-05,
|
| 1628 |
+
"loss": 0.1419,
|
| 1629 |
+
"step": 1080
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 3.3230769230769233,
|
| 1633 |
+
"eval_loss": 0.9528248310089111,
|
| 1634 |
+
"eval_runtime": 1.174,
|
| 1635 |
+
"eval_samples_per_second": 97.107,
|
| 1636 |
+
"eval_steps_per_second": 5.111,
|
| 1637 |
+
"step": 1080
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 3.353846153846154,
|
| 1641 |
+
"grad_norm": 1.7857142686843872,
|
| 1642 |
+
"learning_rate": 1.6810158587867973e-05,
|
| 1643 |
+
"loss": 0.128,
|
| 1644 |
+
"step": 1090
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 3.353846153846154,
|
| 1648 |
+
"eval_loss": 0.9331367611885071,
|
| 1649 |
+
"eval_runtime": 1.1718,
|
| 1650 |
+
"eval_samples_per_second": 97.282,
|
| 1651 |
+
"eval_steps_per_second": 5.12,
|
| 1652 |
+
"step": 1090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 3.3846153846153846,
|
| 1656 |
+
"grad_norm": 1.5178117752075195,
|
| 1657 |
+
"learning_rate": 1.67311180742757e-05,
|
| 1658 |
+
"loss": 0.112,
|
| 1659 |
+
"step": 1100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 3.3846153846153846,
|
| 1663 |
+
"eval_loss": 0.9547919631004333,
|
| 1664 |
+
"eval_runtime": 1.1695,
|
| 1665 |
+
"eval_samples_per_second": 97.48,
|
| 1666 |
+
"eval_steps_per_second": 5.131,
|
| 1667 |
+
"step": 1100
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 3.4153846153846152,
|
| 1671 |
+
"grad_norm": 1.3151490688323975,
|
| 1672 |
+
"learning_rate": 1.665130107931666e-05,
|
| 1673 |
+
"loss": 0.1119,
|
| 1674 |
+
"step": 1110
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 3.4153846153846152,
|
| 1678 |
+
"eval_loss": 0.9812674522399902,
|
| 1679 |
+
"eval_runtime": 1.1666,
|
| 1680 |
+
"eval_samples_per_second": 97.721,
|
| 1681 |
+
"eval_steps_per_second": 5.143,
|
| 1682 |
+
"step": 1110
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 3.4461538461538463,
|
| 1686 |
+
"grad_norm": 1.4334938526153564,
|
| 1687 |
+
"learning_rate": 1.657071681043731e-05,
|
| 1688 |
+
"loss": 0.1042,
|
| 1689 |
+
"step": 1120
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 3.4461538461538463,
|
| 1693 |
+
"eval_loss": 0.9601649641990662,
|
| 1694 |
+
"eval_runtime": 1.1706,
|
| 1695 |
+
"eval_samples_per_second": 97.388,
|
| 1696 |
+
"eval_steps_per_second": 5.126,
|
| 1697 |
+
"step": 1120
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 3.476923076923077,
|
| 1701 |
+
"grad_norm": 1.3328880071640015,
|
| 1702 |
+
"learning_rate": 1.648937456359451e-05,
|
| 1703 |
+
"loss": 0.1129,
|
| 1704 |
+
"step": 1130
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 3.476923076923077,
|
| 1708 |
+
"eval_loss": 0.9889415502548218,
|
| 1709 |
+
"eval_runtime": 1.1689,
|
| 1710 |
+
"eval_samples_per_second": 97.527,
|
| 1711 |
+
"eval_steps_per_second": 5.133,
|
| 1712 |
+
"step": 1130
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 3.5076923076923077,
|
| 1716 |
+
"grad_norm": 1.8561350107192993,
|
| 1717 |
+
"learning_rate": 1.640728372218317e-05,
|
| 1718 |
+
"loss": 0.1247,
|
| 1719 |
+
"step": 1140
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 3.5076923076923077,
|
| 1723 |
+
"eval_loss": 0.9682590961456299,
|
| 1724 |
+
"eval_runtime": 1.1697,
|
| 1725 |
+
"eval_samples_per_second": 97.461,
|
| 1726 |
+
"eval_steps_per_second": 5.13,
|
| 1727 |
+
"step": 1140
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 3.5384615384615383,
|
| 1731 |
+
"grad_norm": 1.1813424825668335,
|
| 1732 |
+
"learning_rate": 1.6324453755953772e-05,
|
| 1733 |
+
"loss": 0.1213,
|
| 1734 |
+
"step": 1150
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 3.5384615384615383,
|
| 1738 |
+
"eval_loss": 0.9461174607276917,
|
| 1739 |
+
"eval_runtime": 1.1671,
|
| 1740 |
+
"eval_samples_per_second": 97.674,
|
| 1741 |
+
"eval_steps_per_second": 5.141,
|
| 1742 |
+
"step": 1150
|
| 1743 |
+
}
|
| 1744 |
+
],
|
| 1745 |
+
"logging_steps": 10,
|
| 1746 |
+
"max_steps": 3250,
|
| 1747 |
+
"num_input_tokens_seen": 0,
|
| 1748 |
+
"num_train_epochs": 10,
|
| 1749 |
+
"save_steps": 50,
|
| 1750 |
+
"total_flos": 7.177114260340736e+16,
|
| 1751 |
+
"train_batch_size": 2,
|
| 1752 |
+
"trial_name": null,
|
| 1753 |
+
"trial_params": null
|
| 1754 |
+
}
|
{checkpoint-250 → checkpoint-1150}/training_args.bin
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/added_tokens.json
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/config.json
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/generation_config.json
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/model.safetensors.index.json
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/special_tokens_map.json
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/tokenizer.json
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/tokenizer.model
RENAMED
|
File without changes
|
{checkpoint-300 → checkpoint-1200}/tokenizer_config.json
RENAMED
|
File without changes
|
checkpoint-1200/trainer_state.json
ADDED
|
@@ -0,0 +1,1829 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.6923076923076925,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 1200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.006153846153846154,
|
| 13 |
+
"eval_loss": 1.487821340560913,
|
| 14 |
+
"eval_runtime": 1.2374,
|
| 15 |
+
"eval_samples_per_second": 92.131,
|
| 16 |
+
"eval_steps_per_second": 4.849,
|
| 17 |
+
"step": 2
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.03076923076923077,
|
| 21 |
+
"grad_norm": 20.007999420166016,
|
| 22 |
+
"learning_rate": 6.153846153846155e-07,
|
| 23 |
+
"loss": 1.4703,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.03076923076923077,
|
| 28 |
+
"eval_loss": 1.4221965074539185,
|
| 29 |
+
"eval_runtime": 1.1701,
|
| 30 |
+
"eval_samples_per_second": 97.431,
|
| 31 |
+
"eval_steps_per_second": 5.128,
|
| 32 |
+
"step": 10
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06153846153846154,
|
| 36 |
+
"grad_norm": 29.657033920288086,
|
| 37 |
+
"learning_rate": 1.230769230769231e-06,
|
| 38 |
+
"loss": 1.4402,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.06153846153846154,
|
| 43 |
+
"eval_loss": 1.238189935684204,
|
| 44 |
+
"eval_runtime": 1.1729,
|
| 45 |
+
"eval_samples_per_second": 97.195,
|
| 46 |
+
"eval_steps_per_second": 5.116,
|
| 47 |
+
"step": 20
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.09230769230769231,
|
| 51 |
+
"grad_norm": 10.80688190460205,
|
| 52 |
+
"learning_rate": 1.8461538461538465e-06,
|
| 53 |
+
"loss": 1.0957,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09230769230769231,
|
| 58 |
+
"eval_loss": 0.908875048160553,
|
| 59 |
+
"eval_runtime": 1.1739,
|
| 60 |
+
"eval_samples_per_second": 97.111,
|
| 61 |
+
"eval_steps_per_second": 5.111,
|
| 62 |
+
"step": 30
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.12307692307692308,
|
| 66 |
+
"grad_norm": 12.914667129516602,
|
| 67 |
+
"learning_rate": 2.461538461538462e-06,
|
| 68 |
+
"loss": 0.8136,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.12307692307692308,
|
| 73 |
+
"eval_loss": 0.7436326146125793,
|
| 74 |
+
"eval_runtime": 1.1712,
|
| 75 |
+
"eval_samples_per_second": 97.34,
|
| 76 |
+
"eval_steps_per_second": 5.123,
|
| 77 |
+
"step": 40
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.15384615384615385,
|
| 81 |
+
"grad_norm": 5.206438064575195,
|
| 82 |
+
"learning_rate": 3.0769230769230774e-06,
|
| 83 |
+
"loss": 0.7864,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.15384615384615385,
|
| 88 |
+
"eval_loss": 0.6924600601196289,
|
| 89 |
+
"eval_runtime": 1.1718,
|
| 90 |
+
"eval_samples_per_second": 97.287,
|
| 91 |
+
"eval_steps_per_second": 5.12,
|
| 92 |
+
"step": 50
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.18461538461538463,
|
| 96 |
+
"grad_norm": 5.670599460601807,
|
| 97 |
+
"learning_rate": 3.692307692307693e-06,
|
| 98 |
+
"loss": 0.622,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.18461538461538463,
|
| 103 |
+
"eval_loss": 0.6791077852249146,
|
| 104 |
+
"eval_runtime": 1.1678,
|
| 105 |
+
"eval_samples_per_second": 97.62,
|
| 106 |
+
"eval_steps_per_second": 5.138,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.2153846153846154,
|
| 111 |
+
"grad_norm": 3.6266868114471436,
|
| 112 |
+
"learning_rate": 4.307692307692308e-06,
|
| 113 |
+
"loss": 0.6672,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2153846153846154,
|
| 118 |
+
"eval_loss": 0.6631948947906494,
|
| 119 |
+
"eval_runtime": 1.1626,
|
| 120 |
+
"eval_samples_per_second": 98.06,
|
| 121 |
+
"eval_steps_per_second": 5.161,
|
| 122 |
+
"step": 70
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.24615384615384617,
|
| 126 |
+
"grad_norm": 3.1027884483337402,
|
| 127 |
+
"learning_rate": 4.923076923076924e-06,
|
| 128 |
+
"loss": 0.6097,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.24615384615384617,
|
| 133 |
+
"eval_loss": 0.6625475883483887,
|
| 134 |
+
"eval_runtime": 1.1717,
|
| 135 |
+
"eval_samples_per_second": 97.294,
|
| 136 |
+
"eval_steps_per_second": 5.121,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.27692307692307694,
|
| 141 |
+
"grad_norm": 3.233839511871338,
|
| 142 |
+
"learning_rate": 5.538461538461539e-06,
|
| 143 |
+
"loss": 0.6957,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.27692307692307694,
|
| 148 |
+
"eval_loss": 0.6590337157249451,
|
| 149 |
+
"eval_runtime": 1.1664,
|
| 150 |
+
"eval_samples_per_second": 97.738,
|
| 151 |
+
"eval_steps_per_second": 5.144,
|
| 152 |
+
"step": 90
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.3076923076923077,
|
| 156 |
+
"grad_norm": 4.932614803314209,
|
| 157 |
+
"learning_rate": 6.153846153846155e-06,
|
| 158 |
+
"loss": 0.6994,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.3076923076923077,
|
| 163 |
+
"eval_loss": 0.6592673659324646,
|
| 164 |
+
"eval_runtime": 1.1692,
|
| 165 |
+
"eval_samples_per_second": 97.502,
|
| 166 |
+
"eval_steps_per_second": 5.132,
|
| 167 |
+
"step": 100
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.3384615384615385,
|
| 171 |
+
"grad_norm": 2.642303466796875,
|
| 172 |
+
"learning_rate": 6.76923076923077e-06,
|
| 173 |
+
"loss": 0.6501,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.3384615384615385,
|
| 178 |
+
"eval_loss": 0.6589992642402649,
|
| 179 |
+
"eval_runtime": 1.1669,
|
| 180 |
+
"eval_samples_per_second": 97.691,
|
| 181 |
+
"eval_steps_per_second": 5.142,
|
| 182 |
+
"step": 110
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.36923076923076925,
|
| 186 |
+
"grad_norm": 3.45409893989563,
|
| 187 |
+
"learning_rate": 7.384615384615386e-06,
|
| 188 |
+
"loss": 0.7512,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.36923076923076925,
|
| 193 |
+
"eval_loss": 0.6568591594696045,
|
| 194 |
+
"eval_runtime": 1.1778,
|
| 195 |
+
"eval_samples_per_second": 96.79,
|
| 196 |
+
"eval_steps_per_second": 5.094,
|
| 197 |
+
"step": 120
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.4,
|
| 201 |
+
"grad_norm": 8.460351943969727,
|
| 202 |
+
"learning_rate": 8.000000000000001e-06,
|
| 203 |
+
"loss": 0.5996,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.4,
|
| 208 |
+
"eval_loss": 0.6578297019004822,
|
| 209 |
+
"eval_runtime": 1.1683,
|
| 210 |
+
"eval_samples_per_second": 97.575,
|
| 211 |
+
"eval_steps_per_second": 5.136,
|
| 212 |
+
"step": 130
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4307692307692308,
|
| 216 |
+
"grad_norm": 2.4495465755462646,
|
| 217 |
+
"learning_rate": 8.615384615384617e-06,
|
| 218 |
+
"loss": 0.6498,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4307692307692308,
|
| 223 |
+
"eval_loss": 0.6588199138641357,
|
| 224 |
+
"eval_runtime": 1.1664,
|
| 225 |
+
"eval_samples_per_second": 97.739,
|
| 226 |
+
"eval_steps_per_second": 5.144,
|
| 227 |
+
"step": 140
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.46153846153846156,
|
| 231 |
+
"grad_norm": 2.6779162883758545,
|
| 232 |
+
"learning_rate": 9.230769230769232e-06,
|
| 233 |
+
"loss": 0.703,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.46153846153846156,
|
| 238 |
+
"eval_loss": 0.6626676321029663,
|
| 239 |
+
"eval_runtime": 1.1764,
|
| 240 |
+
"eval_samples_per_second": 96.904,
|
| 241 |
+
"eval_steps_per_second": 5.1,
|
| 242 |
+
"step": 150
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.49230769230769234,
|
| 246 |
+
"grad_norm": 8.147880554199219,
|
| 247 |
+
"learning_rate": 9.846153846153848e-06,
|
| 248 |
+
"loss": 0.6649,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.49230769230769234,
|
| 253 |
+
"eval_loss": 0.6644802689552307,
|
| 254 |
+
"eval_runtime": 1.1701,
|
| 255 |
+
"eval_samples_per_second": 97.424,
|
| 256 |
+
"eval_steps_per_second": 5.128,
|
| 257 |
+
"step": 160
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.5230769230769231,
|
| 261 |
+
"grad_norm": 2.7410728931427,
|
| 262 |
+
"learning_rate": 1.0461538461538463e-05,
|
| 263 |
+
"loss": 0.6704,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.5230769230769231,
|
| 268 |
+
"eval_loss": 0.6591455340385437,
|
| 269 |
+
"eval_runtime": 1.1722,
|
| 270 |
+
"eval_samples_per_second": 97.253,
|
| 271 |
+
"eval_steps_per_second": 5.119,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.5538461538461539,
|
| 276 |
+
"grad_norm": 6.886598587036133,
|
| 277 |
+
"learning_rate": 1.1076923076923079e-05,
|
| 278 |
+
"loss": 0.6073,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.5538461538461539,
|
| 283 |
+
"eval_loss": 0.6617754697799683,
|
| 284 |
+
"eval_runtime": 1.1682,
|
| 285 |
+
"eval_samples_per_second": 97.588,
|
| 286 |
+
"eval_steps_per_second": 5.136,
|
| 287 |
+
"step": 180
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.5846153846153846,
|
| 291 |
+
"grad_norm": 4.26249361038208,
|
| 292 |
+
"learning_rate": 1.1692307692307694e-05,
|
| 293 |
+
"loss": 0.7556,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.5846153846153846,
|
| 298 |
+
"eval_loss": 0.6659030318260193,
|
| 299 |
+
"eval_runtime": 1.1721,
|
| 300 |
+
"eval_samples_per_second": 97.265,
|
| 301 |
+
"eval_steps_per_second": 5.119,
|
| 302 |
+
"step": 190
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.6153846153846154,
|
| 306 |
+
"grad_norm": 4.204222202301025,
|
| 307 |
+
"learning_rate": 1.230769230769231e-05,
|
| 308 |
+
"loss": 0.6583,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.6153846153846154,
|
| 313 |
+
"eval_loss": 0.6660778522491455,
|
| 314 |
+
"eval_runtime": 1.1728,
|
| 315 |
+
"eval_samples_per_second": 97.201,
|
| 316 |
+
"eval_steps_per_second": 5.116,
|
| 317 |
+
"step": 200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6461538461538462,
|
| 321 |
+
"grad_norm": 3.154658079147339,
|
| 322 |
+
"learning_rate": 1.2923076923076925e-05,
|
| 323 |
+
"loss": 0.7288,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6461538461538462,
|
| 328 |
+
"eval_loss": 0.6642422676086426,
|
| 329 |
+
"eval_runtime": 1.166,
|
| 330 |
+
"eval_samples_per_second": 97.772,
|
| 331 |
+
"eval_steps_per_second": 5.146,
|
| 332 |
+
"step": 210
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.676923076923077,
|
| 336 |
+
"grad_norm": 3.248359203338623,
|
| 337 |
+
"learning_rate": 1.353846153846154e-05,
|
| 338 |
+
"loss": 0.6702,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.676923076923077,
|
| 343 |
+
"eval_loss": 0.6680281758308411,
|
| 344 |
+
"eval_runtime": 1.1691,
|
| 345 |
+
"eval_samples_per_second": 97.507,
|
| 346 |
+
"eval_steps_per_second": 5.132,
|
| 347 |
+
"step": 220
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.7076923076923077,
|
| 351 |
+
"grad_norm": 2.5507216453552246,
|
| 352 |
+
"learning_rate": 1.4153846153846156e-05,
|
| 353 |
+
"loss": 0.66,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.7076923076923077,
|
| 358 |
+
"eval_loss": 0.6713310480117798,
|
| 359 |
+
"eval_runtime": 1.167,
|
| 360 |
+
"eval_samples_per_second": 97.683,
|
| 361 |
+
"eval_steps_per_second": 5.141,
|
| 362 |
+
"step": 230
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.7384615384615385,
|
| 366 |
+
"grad_norm": 2.1180007457733154,
|
| 367 |
+
"learning_rate": 1.4769230769230772e-05,
|
| 368 |
+
"loss": 0.6172,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.7384615384615385,
|
| 373 |
+
"eval_loss": 0.6759226322174072,
|
| 374 |
+
"eval_runtime": 1.1758,
|
| 375 |
+
"eval_samples_per_second": 96.954,
|
| 376 |
+
"eval_steps_per_second": 5.103,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.7692307692307693,
|
| 381 |
+
"grad_norm": 2.8318371772766113,
|
| 382 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 383 |
+
"loss": 0.5954,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.7692307692307693,
|
| 388 |
+
"eval_loss": 0.6854314208030701,
|
| 389 |
+
"eval_runtime": 1.1698,
|
| 390 |
+
"eval_samples_per_second": 97.457,
|
| 391 |
+
"eval_steps_per_second": 5.129,
|
| 392 |
+
"step": 250
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.8,
|
| 396 |
+
"grad_norm": 3.519829034805298,
|
| 397 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 398 |
+
"loss": 0.6639,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.8,
|
| 403 |
+
"eval_loss": 0.6849851012229919,
|
| 404 |
+
"eval_runtime": 1.1732,
|
| 405 |
+
"eval_samples_per_second": 97.173,
|
| 406 |
+
"eval_steps_per_second": 5.114,
|
| 407 |
+
"step": 260
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.8307692307692308,
|
| 411 |
+
"grad_norm": 3.7909388542175293,
|
| 412 |
+
"learning_rate": 1.6615384615384618e-05,
|
| 413 |
+
"loss": 0.6892,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.8307692307692308,
|
| 418 |
+
"eval_loss": 0.6863855123519897,
|
| 419 |
+
"eval_runtime": 1.1689,
|
| 420 |
+
"eval_samples_per_second": 97.53,
|
| 421 |
+
"eval_steps_per_second": 5.133,
|
| 422 |
+
"step": 270
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8615384615384616,
|
| 426 |
+
"grad_norm": 8.278326988220215,
|
| 427 |
+
"learning_rate": 1.7230769230769234e-05,
|
| 428 |
+
"loss": 0.6014,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8615384615384616,
|
| 433 |
+
"eval_loss": 0.6890373826026917,
|
| 434 |
+
"eval_runtime": 1.1722,
|
| 435 |
+
"eval_samples_per_second": 97.251,
|
| 436 |
+
"eval_steps_per_second": 5.118,
|
| 437 |
+
"step": 280
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.8923076923076924,
|
| 441 |
+
"grad_norm": 2.616001605987549,
|
| 442 |
+
"learning_rate": 1.784615384615385e-05,
|
| 443 |
+
"loss": 0.6775,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.8923076923076924,
|
| 448 |
+
"eval_loss": 0.6797110438346863,
|
| 449 |
+
"eval_runtime": 1.1888,
|
| 450 |
+
"eval_samples_per_second": 95.897,
|
| 451 |
+
"eval_steps_per_second": 5.047,
|
| 452 |
+
"step": 290
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.9230769230769231,
|
| 456 |
+
"grad_norm": 2.4450528621673584,
|
| 457 |
+
"learning_rate": 1.8461538461538465e-05,
|
| 458 |
+
"loss": 0.7434,
|
| 459 |
+
"step": 300
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.9230769230769231,
|
| 463 |
+
"eval_loss": 0.6779448986053467,
|
| 464 |
+
"eval_runtime": 1.1722,
|
| 465 |
+
"eval_samples_per_second": 97.254,
|
| 466 |
+
"eval_steps_per_second": 5.119,
|
| 467 |
+
"step": 300
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.9538461538461539,
|
| 471 |
+
"grad_norm": 4.121164798736572,
|
| 472 |
+
"learning_rate": 1.907692307692308e-05,
|
| 473 |
+
"loss": 0.6413,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.9538461538461539,
|
| 478 |
+
"eval_loss": 0.6819297671318054,
|
| 479 |
+
"eval_runtime": 1.1683,
|
| 480 |
+
"eval_samples_per_second": 97.574,
|
| 481 |
+
"eval_steps_per_second": 5.135,
|
| 482 |
+
"step": 310
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.9846153846153847,
|
| 486 |
+
"grad_norm": 2.057229995727539,
|
| 487 |
+
"learning_rate": 1.9692307692307696e-05,
|
| 488 |
+
"loss": 0.6511,
|
| 489 |
+
"step": 320
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.9846153846153847,
|
| 493 |
+
"eval_loss": 0.6861481666564941,
|
| 494 |
+
"eval_runtime": 1.1699,
|
| 495 |
+
"eval_samples_per_second": 97.444,
|
| 496 |
+
"eval_steps_per_second": 5.129,
|
| 497 |
+
"step": 320
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 1.0153846153846153,
|
| 501 |
+
"grad_norm": 1.7295210361480713,
|
| 502 |
+
"learning_rate": 1.9999855802751384e-05,
|
| 503 |
+
"loss": 0.5872,
|
| 504 |
+
"step": 330
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.0153846153846153,
|
| 508 |
+
"eval_loss": 0.7481781244277954,
|
| 509 |
+
"eval_runtime": 1.172,
|
| 510 |
+
"eval_samples_per_second": 97.273,
|
| 511 |
+
"eval_steps_per_second": 5.12,
|
| 512 |
+
"step": 330
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.0461538461538462,
|
| 516 |
+
"grad_norm": 3.6910336017608643,
|
| 517 |
+
"learning_rate": 1.9998702249713747e-05,
|
| 518 |
+
"loss": 0.438,
|
| 519 |
+
"step": 340
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.0461538461538462,
|
| 523 |
+
"eval_loss": 0.7471169829368591,
|
| 524 |
+
"eval_runtime": 1.175,
|
| 525 |
+
"eval_samples_per_second": 97.019,
|
| 526 |
+
"eval_steps_per_second": 5.106,
|
| 527 |
+
"step": 340
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.0769230769230769,
|
| 531 |
+
"grad_norm": 3.9220831394195557,
|
| 532 |
+
"learning_rate": 1.9996395276708856e-05,
|
| 533 |
+
"loss": 0.4141,
|
| 534 |
+
"step": 350
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0769230769230769,
|
| 538 |
+
"eval_loss": 0.7425897121429443,
|
| 539 |
+
"eval_runtime": 1.1888,
|
| 540 |
+
"eval_samples_per_second": 95.893,
|
| 541 |
+
"eval_steps_per_second": 5.047,
|
| 542 |
+
"step": 350
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.1076923076923078,
|
| 546 |
+
"grad_norm": 2.1136879920959473,
|
| 547 |
+
"learning_rate": 1.9992935149862116e-05,
|
| 548 |
+
"loss": 0.4173,
|
| 549 |
+
"step": 360
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.1076923076923078,
|
| 553 |
+
"eval_loss": 0.7477542161941528,
|
| 554 |
+
"eval_runtime": 1.1705,
|
| 555 |
+
"eval_samples_per_second": 97.398,
|
| 556 |
+
"eval_steps_per_second": 5.126,
|
| 557 |
+
"step": 360
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.1384615384615384,
|
| 561 |
+
"grad_norm": 2.187299966812134,
|
| 562 |
+
"learning_rate": 1.998832226832327e-05,
|
| 563 |
+
"loss": 0.4242,
|
| 564 |
+
"step": 370
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.1384615384615384,
|
| 568 |
+
"eval_loss": 0.7475994229316711,
|
| 569 |
+
"eval_runtime": 1.1683,
|
| 570 |
+
"eval_samples_per_second": 97.582,
|
| 571 |
+
"eval_steps_per_second": 5.136,
|
| 572 |
+
"step": 370
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 1.1692307692307693,
|
| 576 |
+
"grad_norm": 2.683750629425049,
|
| 577 |
+
"learning_rate": 1.9982557164220335e-05,
|
| 578 |
+
"loss": 0.4413,
|
| 579 |
+
"step": 380
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.1692307692307693,
|
| 583 |
+
"eval_loss": 0.7552060484886169,
|
| 584 |
+
"eval_runtime": 1.1689,
|
| 585 |
+
"eval_samples_per_second": 97.529,
|
| 586 |
+
"eval_steps_per_second": 5.133,
|
| 587 |
+
"step": 380
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 1.2,
|
| 591 |
+
"grad_norm": 3.312657117843628,
|
| 592 |
+
"learning_rate": 1.9975640502598243e-05,
|
| 593 |
+
"loss": 0.3361,
|
| 594 |
+
"step": 390
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.2,
|
| 598 |
+
"eval_loss": 0.7630029916763306,
|
| 599 |
+
"eval_runtime": 1.1689,
|
| 600 |
+
"eval_samples_per_second": 97.529,
|
| 601 |
+
"eval_steps_per_second": 5.133,
|
| 602 |
+
"step": 390
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 1.2307692307692308,
|
| 606 |
+
"grad_norm": 2.3290536403656006,
|
| 607 |
+
"learning_rate": 1.9967573081342103e-05,
|
| 608 |
+
"loss": 0.4174,
|
| 609 |
+
"step": 400
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.2307692307692308,
|
| 613 |
+
"eval_loss": 0.7669293880462646,
|
| 614 |
+
"eval_runtime": 1.1711,
|
| 615 |
+
"eval_samples_per_second": 97.345,
|
| 616 |
+
"eval_steps_per_second": 5.123,
|
| 617 |
+
"step": 400
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.2615384615384615,
|
| 621 |
+
"grad_norm": 2.4483697414398193,
|
| 622 |
+
"learning_rate": 1.9958355831085155e-05,
|
| 623 |
+
"loss": 0.4045,
|
| 624 |
+
"step": 410
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 1.2615384615384615,
|
| 628 |
+
"eval_loss": 0.7652024626731873,
|
| 629 |
+
"eval_runtime": 1.1697,
|
| 630 |
+
"eval_samples_per_second": 97.461,
|
| 631 |
+
"eval_steps_per_second": 5.13,
|
| 632 |
+
"step": 410
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.2923076923076924,
|
| 636 |
+
"grad_norm": 1.703616976737976,
|
| 637 |
+
"learning_rate": 1.9947989815101444e-05,
|
| 638 |
+
"loss": 0.4292,
|
| 639 |
+
"step": 420
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.2923076923076924,
|
| 643 |
+
"eval_loss": 0.7553891539573669,
|
| 644 |
+
"eval_runtime": 1.1719,
|
| 645 |
+
"eval_samples_per_second": 97.277,
|
| 646 |
+
"eval_steps_per_second": 5.12,
|
| 647 |
+
"step": 420
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.323076923076923,
|
| 651 |
+
"grad_norm": 1.6245399713516235,
|
| 652 |
+
"learning_rate": 1.9936476229183133e-05,
|
| 653 |
+
"loss": 0.4164,
|
| 654 |
+
"step": 430
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.323076923076923,
|
| 658 |
+
"eval_loss": 0.7711736559867859,
|
| 659 |
+
"eval_runtime": 1.1776,
|
| 660 |
+
"eval_samples_per_second": 96.806,
|
| 661 |
+
"eval_steps_per_second": 5.095,
|
| 662 |
+
"step": 430
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 1.353846153846154,
|
| 666 |
+
"grad_norm": 1.9511611461639404,
|
| 667 |
+
"learning_rate": 1.992381640150257e-05,
|
| 668 |
+
"loss": 0.4337,
|
| 669 |
+
"step": 440
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.353846153846154,
|
| 673 |
+
"eval_loss": 0.7438372373580933,
|
| 674 |
+
"eval_runtime": 1.1692,
|
| 675 |
+
"eval_samples_per_second": 97.5,
|
| 676 |
+
"eval_steps_per_second": 5.132,
|
| 677 |
+
"step": 440
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 1.3846153846153846,
|
| 681 |
+
"grad_norm": 2.466737747192383,
|
| 682 |
+
"learning_rate": 1.9910011792459086e-05,
|
| 683 |
+
"loss": 0.3712,
|
| 684 |
+
"step": 450
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 1.3846153846153846,
|
| 688 |
+
"eval_loss": 0.8045614361763,
|
| 689 |
+
"eval_runtime": 1.1689,
|
| 690 |
+
"eval_samples_per_second": 97.53,
|
| 691 |
+
"eval_steps_per_second": 5.133,
|
| 692 |
+
"step": 450
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 1.4153846153846155,
|
| 696 |
+
"grad_norm": 2.473851203918457,
|
| 697 |
+
"learning_rate": 1.9895063994510512e-05,
|
| 698 |
+
"loss": 0.516,
|
| 699 |
+
"step": 460
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 1.4153846153846155,
|
| 703 |
+
"eval_loss": 0.7463352084159851,
|
| 704 |
+
"eval_runtime": 1.1697,
|
| 705 |
+
"eval_samples_per_second": 97.464,
|
| 706 |
+
"eval_steps_per_second": 5.13,
|
| 707 |
+
"step": 460
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 1.4461538461538461,
|
| 711 |
+
"grad_norm": 2.757955312728882,
|
| 712 |
+
"learning_rate": 1.9878974731989487e-05,
|
| 713 |
+
"loss": 0.4965,
|
| 714 |
+
"step": 470
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 1.4461538461538461,
|
| 718 |
+
"eval_loss": 0.7571713924407959,
|
| 719 |
+
"eval_runtime": 1.1688,
|
| 720 |
+
"eval_samples_per_second": 97.534,
|
| 721 |
+
"eval_steps_per_second": 5.133,
|
| 722 |
+
"step": 470
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.476923076923077,
|
| 726 |
+
"grad_norm": 3.9884915351867676,
|
| 727 |
+
"learning_rate": 1.9861745860904538e-05,
|
| 728 |
+
"loss": 0.3973,
|
| 729 |
+
"step": 480
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.476923076923077,
|
| 733 |
+
"eval_loss": 0.7763081789016724,
|
| 734 |
+
"eval_runtime": 1.168,
|
| 735 |
+
"eval_samples_per_second": 97.605,
|
| 736 |
+
"eval_steps_per_second": 5.137,
|
| 737 |
+
"step": 480
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5076923076923077,
|
| 741 |
+
"grad_norm": 1.9474631547927856,
|
| 742 |
+
"learning_rate": 1.9843379368725978e-05,
|
| 743 |
+
"loss": 0.4625,
|
| 744 |
+
"step": 490
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5076923076923077,
|
| 748 |
+
"eval_loss": 0.7595367431640625,
|
| 749 |
+
"eval_runtime": 1.1708,
|
| 750 |
+
"eval_samples_per_second": 97.37,
|
| 751 |
+
"eval_steps_per_second": 5.125,
|
| 752 |
+
"step": 490
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.5384615384615383,
|
| 756 |
+
"grad_norm": 2.2907869815826416,
|
| 757 |
+
"learning_rate": 1.9823877374156647e-05,
|
| 758 |
+
"loss": 0.4371,
|
| 759 |
+
"step": 500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.5384615384615383,
|
| 763 |
+
"eval_loss": 0.762865424156189,
|
| 764 |
+
"eval_runtime": 1.1766,
|
| 765 |
+
"eval_samples_per_second": 96.891,
|
| 766 |
+
"eval_steps_per_second": 5.1,
|
| 767 |
+
"step": 500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.5692307692307692,
|
| 771 |
+
"grad_norm": 3.067608594894409,
|
| 772 |
+
"learning_rate": 1.9803242126887496e-05,
|
| 773 |
+
"loss": 0.45,
|
| 774 |
+
"step": 510
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.5692307692307692,
|
| 778 |
+
"eval_loss": 0.7638718485832214,
|
| 779 |
+
"eval_runtime": 1.1684,
|
| 780 |
+
"eval_samples_per_second": 97.568,
|
| 781 |
+
"eval_steps_per_second": 5.135,
|
| 782 |
+
"step": 510
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 1.6,
|
| 786 |
+
"grad_norm": 2.1113197803497314,
|
| 787 |
+
"learning_rate": 1.9781476007338058e-05,
|
| 788 |
+
"loss": 0.4524,
|
| 789 |
+
"step": 520
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.6,
|
| 793 |
+
"eval_loss": 0.7560412883758545,
|
| 794 |
+
"eval_runtime": 1.1719,
|
| 795 |
+
"eval_samples_per_second": 97.278,
|
| 796 |
+
"eval_steps_per_second": 5.12,
|
| 797 |
+
"step": 520
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 1.6307692307692307,
|
| 801 |
+
"grad_norm": 1.889585256576538,
|
| 802 |
+
"learning_rate": 1.9758581526381878e-05,
|
| 803 |
+
"loss": 0.4287,
|
| 804 |
+
"step": 530
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.6307692307692307,
|
| 808 |
+
"eval_loss": 0.7591350674629211,
|
| 809 |
+
"eval_runtime": 1.1693,
|
| 810 |
+
"eval_samples_per_second": 97.495,
|
| 811 |
+
"eval_steps_per_second": 5.131,
|
| 812 |
+
"step": 530
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 1.6615384615384614,
|
| 816 |
+
"grad_norm": 2.519418716430664,
|
| 817 |
+
"learning_rate": 1.973456132505684e-05,
|
| 818 |
+
"loss": 0.4603,
|
| 819 |
+
"step": 540
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 1.6615384615384614,
|
| 823 |
+
"eval_loss": 0.7536805868148804,
|
| 824 |
+
"eval_runtime": 1.1799,
|
| 825 |
+
"eval_samples_per_second": 96.617,
|
| 826 |
+
"eval_steps_per_second": 5.085,
|
| 827 |
+
"step": 540
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.6923076923076923,
|
| 831 |
+
"grad_norm": 2.7455596923828125,
|
| 832 |
+
"learning_rate": 1.9709418174260523e-05,
|
| 833 |
+
"loss": 0.4293,
|
| 834 |
+
"step": 550
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.6923076923076923,
|
| 838 |
+
"eval_loss": 0.763169527053833,
|
| 839 |
+
"eval_runtime": 1.1696,
|
| 840 |
+
"eval_samples_per_second": 97.47,
|
| 841 |
+
"eval_steps_per_second": 5.13,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7230769230769232,
|
| 846 |
+
"grad_norm": 1.9823411703109741,
|
| 847 |
+
"learning_rate": 1.9683154974430544e-05,
|
| 848 |
+
"loss": 0.4643,
|
| 849 |
+
"step": 560
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7230769230769232,
|
| 853 |
+
"eval_loss": 0.7607314586639404,
|
| 854 |
+
"eval_runtime": 1.1712,
|
| 855 |
+
"eval_samples_per_second": 97.336,
|
| 856 |
+
"eval_steps_per_second": 5.123,
|
| 857 |
+
"step": 560
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.7538461538461538,
|
| 861 |
+
"grad_norm": 2.2437684535980225,
|
| 862 |
+
"learning_rate": 1.965577475520999e-05,
|
| 863 |
+
"loss": 0.452,
|
| 864 |
+
"step": 570
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.7538461538461538,
|
| 868 |
+
"eval_loss": 0.7615717649459839,
|
| 869 |
+
"eval_runtime": 1.1946,
|
| 870 |
+
"eval_samples_per_second": 95.425,
|
| 871 |
+
"eval_steps_per_second": 5.022,
|
| 872 |
+
"step": 570
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 1.7846153846153845,
|
| 876 |
+
"grad_norm": 3.0740182399749756,
|
| 877 |
+
"learning_rate": 1.962728067509791e-05,
|
| 878 |
+
"loss": 0.4885,
|
| 879 |
+
"step": 580
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 1.7846153846153845,
|
| 883 |
+
"eval_loss": 0.7764689326286316,
|
| 884 |
+
"eval_runtime": 1.1714,
|
| 885 |
+
"eval_samples_per_second": 97.32,
|
| 886 |
+
"eval_steps_per_second": 5.122,
|
| 887 |
+
"step": 580
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 1.8153846153846154,
|
| 891 |
+
"grad_norm": 2.050313711166382,
|
| 892 |
+
"learning_rate": 1.9597676021084962e-05,
|
| 893 |
+
"loss": 0.4992,
|
| 894 |
+
"step": 590
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 1.8153846153846154,
|
| 898 |
+
"eval_loss": 0.7457339763641357,
|
| 899 |
+
"eval_runtime": 1.1747,
|
| 900 |
+
"eval_samples_per_second": 97.045,
|
| 901 |
+
"eval_steps_per_second": 5.108,
|
| 902 |
+
"step": 590
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.8461538461538463,
|
| 906 |
+
"grad_norm": 2.349470853805542,
|
| 907 |
+
"learning_rate": 1.9566964208274254e-05,
|
| 908 |
+
"loss": 0.3963,
|
| 909 |
+
"step": 600
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 1.8461538461538463,
|
| 913 |
+
"eval_loss": 0.7712545990943909,
|
| 914 |
+
"eval_runtime": 1.17,
|
| 915 |
+
"eval_samples_per_second": 97.433,
|
| 916 |
+
"eval_steps_per_second": 5.128,
|
| 917 |
+
"step": 600
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 1.876923076923077,
|
| 921 |
+
"grad_norm": 1.947912573814392,
|
| 922 |
+
"learning_rate": 1.9535148779487365e-05,
|
| 923 |
+
"loss": 0.4388,
|
| 924 |
+
"step": 610
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 1.876923076923077,
|
| 928 |
+
"eval_loss": 0.772519052028656,
|
| 929 |
+
"eval_runtime": 1.1717,
|
| 930 |
+
"eval_samples_per_second": 97.296,
|
| 931 |
+
"eval_steps_per_second": 5.121,
|
| 932 |
+
"step": 610
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.9076923076923076,
|
| 936 |
+
"grad_norm": 1.8837225437164307,
|
| 937 |
+
"learning_rate": 1.9502233404855672e-05,
|
| 938 |
+
"loss": 0.3965,
|
| 939 |
+
"step": 620
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.9076923076923076,
|
| 943 |
+
"eval_loss": 0.7594517469406128,
|
| 944 |
+
"eval_runtime": 1.2215,
|
| 945 |
+
"eval_samples_per_second": 93.326,
|
| 946 |
+
"eval_steps_per_second": 4.912,
|
| 947 |
+
"step": 620
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9384615384615385,
|
| 951 |
+
"grad_norm": 1.8398792743682861,
|
| 952 |
+
"learning_rate": 1.946822188139696e-05,
|
| 953 |
+
"loss": 0.3898,
|
| 954 |
+
"step": 630
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.9384615384615385,
|
| 958 |
+
"eval_loss": 0.7748451828956604,
|
| 959 |
+
"eval_runtime": 1.2222,
|
| 960 |
+
"eval_samples_per_second": 93.273,
|
| 961 |
+
"eval_steps_per_second": 4.909,
|
| 962 |
+
"step": 630
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.9692307692307693,
|
| 966 |
+
"grad_norm": 2.3971219062805176,
|
| 967 |
+
"learning_rate": 1.9433118132577432e-05,
|
| 968 |
+
"loss": 0.4253,
|
| 969 |
+
"step": 640
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.9692307692307693,
|
| 973 |
+
"eval_loss": 0.7729252576828003,
|
| 974 |
+
"eval_runtime": 1.1743,
|
| 975 |
+
"eval_samples_per_second": 97.081,
|
| 976 |
+
"eval_steps_per_second": 5.11,
|
| 977 |
+
"step": 640
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.0,
|
| 981 |
+
"grad_norm": 1.657328724861145,
|
| 982 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 983 |
+
"loss": 0.4898,
|
| 984 |
+
"step": 650
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.0,
|
| 988 |
+
"eval_loss": 0.7602188587188721,
|
| 989 |
+
"eval_runtime": 1.1738,
|
| 990 |
+
"eval_samples_per_second": 97.117,
|
| 991 |
+
"eval_steps_per_second": 5.111,
|
| 992 |
+
"step": 650
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 2.0307692307692307,
|
| 996 |
+
"grad_norm": 1.3759773969650269,
|
| 997 |
+
"learning_rate": 1.935965028223259e-05,
|
| 998 |
+
"loss": 0.2182,
|
| 999 |
+
"step": 660
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.0307692307692307,
|
| 1003 |
+
"eval_loss": 0.8499692678451538,
|
| 1004 |
+
"eval_runtime": 1.1719,
|
| 1005 |
+
"eval_samples_per_second": 97.276,
|
| 1006 |
+
"eval_steps_per_second": 5.12,
|
| 1007 |
+
"step": 660
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 2.0615384615384613,
|
| 1011 |
+
"grad_norm": 1.4757074117660522,
|
| 1012 |
+
"learning_rate": 1.932129465573568e-05,
|
| 1013 |
+
"loss": 0.2196,
|
| 1014 |
+
"step": 670
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 2.0615384615384613,
|
| 1018 |
+
"eval_loss": 0.866599977016449,
|
| 1019 |
+
"eval_runtime": 1.1739,
|
| 1020 |
+
"eval_samples_per_second": 97.115,
|
| 1021 |
+
"eval_steps_per_second": 5.111,
|
| 1022 |
+
"step": 670
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 2.0923076923076924,
|
| 1026 |
+
"grad_norm": 1.2500200271606445,
|
| 1027 |
+
"learning_rate": 1.9281863752957095e-05,
|
| 1028 |
+
"loss": 0.1772,
|
| 1029 |
+
"step": 680
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 2.0923076923076924,
|
| 1033 |
+
"eval_loss": 0.8642630577087402,
|
| 1034 |
+
"eval_runtime": 1.1753,
|
| 1035 |
+
"eval_samples_per_second": 96.999,
|
| 1036 |
+
"eval_steps_per_second": 5.105,
|
| 1037 |
+
"step": 680
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.123076923076923,
|
| 1041 |
+
"grad_norm": 2.561828851699829,
|
| 1042 |
+
"learning_rate": 1.92413621225262e-05,
|
| 1043 |
+
"loss": 0.2014,
|
| 1044 |
+
"step": 690
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.123076923076923,
|
| 1048 |
+
"eval_loss": 0.8872668743133545,
|
| 1049 |
+
"eval_runtime": 1.1761,
|
| 1050 |
+
"eval_samples_per_second": 96.933,
|
| 1051 |
+
"eval_steps_per_second": 5.102,
|
| 1052 |
+
"step": 690
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.1538461538461537,
|
| 1056 |
+
"grad_norm": 1.8431521654129028,
|
| 1057 |
+
"learning_rate": 1.9199794436588244e-05,
|
| 1058 |
+
"loss": 0.2245,
|
| 1059 |
+
"step": 700
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.1538461538461537,
|
| 1063 |
+
"eval_loss": 0.8799911737442017,
|
| 1064 |
+
"eval_runtime": 1.1766,
|
| 1065 |
+
"eval_samples_per_second": 96.89,
|
| 1066 |
+
"eval_steps_per_second": 5.099,
|
| 1067 |
+
"step": 700
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 2.184615384615385,
|
| 1071 |
+
"grad_norm": 1.4718531370162964,
|
| 1072 |
+
"learning_rate": 1.915716549026541e-05,
|
| 1073 |
+
"loss": 0.1911,
|
| 1074 |
+
"step": 710
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 2.184615384615385,
|
| 1078 |
+
"eval_loss": 0.8806982636451721,
|
| 1079 |
+
"eval_runtime": 1.1696,
|
| 1080 |
+
"eval_samples_per_second": 97.473,
|
| 1081 |
+
"eval_steps_per_second": 5.13,
|
| 1082 |
+
"step": 710
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 2.2153846153846155,
|
| 1086 |
+
"grad_norm": 2.061520576477051,
|
| 1087 |
+
"learning_rate": 1.9113480201103658e-05,
|
| 1088 |
+
"loss": 0.232,
|
| 1089 |
+
"step": 720
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 2.2153846153846155,
|
| 1093 |
+
"eval_loss": 0.8812103271484375,
|
| 1094 |
+
"eval_runtime": 1.1676,
|
| 1095 |
+
"eval_samples_per_second": 97.636,
|
| 1096 |
+
"eval_steps_per_second": 5.139,
|
| 1097 |
+
"step": 720
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 2.246153846153846,
|
| 1101 |
+
"grad_norm": 2.530285596847534,
|
| 1102 |
+
"learning_rate": 1.9068743608505454e-05,
|
| 1103 |
+
"loss": 0.2232,
|
| 1104 |
+
"step": 730
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 2.246153846153846,
|
| 1108 |
+
"eval_loss": 0.8682798147201538,
|
| 1109 |
+
"eval_runtime": 1.1687,
|
| 1110 |
+
"eval_samples_per_second": 97.545,
|
| 1111 |
+
"eval_steps_per_second": 5.134,
|
| 1112 |
+
"step": 730
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 2.276923076923077,
|
| 1116 |
+
"grad_norm": 1.8433207273483276,
|
| 1117 |
+
"learning_rate": 1.902296087314845e-05,
|
| 1118 |
+
"loss": 0.2315,
|
| 1119 |
+
"step": 740
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 2.276923076923077,
|
| 1123 |
+
"eval_loss": 0.8747490048408508,
|
| 1124 |
+
"eval_runtime": 1.1702,
|
| 1125 |
+
"eval_samples_per_second": 97.421,
|
| 1126 |
+
"eval_steps_per_second": 5.127,
|
| 1127 |
+
"step": 740
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 2.3076923076923075,
|
| 1131 |
+
"grad_norm": 2.1529664993286133,
|
| 1132 |
+
"learning_rate": 1.8976137276390145e-05,
|
| 1133 |
+
"loss": 0.239,
|
| 1134 |
+
"step": 750
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 2.3076923076923075,
|
| 1138 |
+
"eval_loss": 0.8778530359268188,
|
| 1139 |
+
"eval_runtime": 1.1674,
|
| 1140 |
+
"eval_samples_per_second": 97.656,
|
| 1141 |
+
"eval_steps_per_second": 5.14,
|
| 1142 |
+
"step": 750
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.3384615384615386,
|
| 1146 |
+
"grad_norm": 1.839979887008667,
|
| 1147 |
+
"learning_rate": 1.892827821965864e-05,
|
| 1148 |
+
"loss": 0.2228,
|
| 1149 |
+
"step": 760
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.3384615384615386,
|
| 1153 |
+
"eval_loss": 0.8769873976707458,
|
| 1154 |
+
"eval_runtime": 1.1718,
|
| 1155 |
+
"eval_samples_per_second": 97.288,
|
| 1156 |
+
"eval_steps_per_second": 5.12,
|
| 1157 |
+
"step": 760
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.3692307692307693,
|
| 1161 |
+
"grad_norm": 1.3814674615859985,
|
| 1162 |
+
"learning_rate": 1.8879389223829592e-05,
|
| 1163 |
+
"loss": 0.1999,
|
| 1164 |
+
"step": 770
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.3692307692307693,
|
| 1168 |
+
"eval_loss": 0.8731983304023743,
|
| 1169 |
+
"eval_runtime": 1.1723,
|
| 1170 |
+
"eval_samples_per_second": 97.242,
|
| 1171 |
+
"eval_steps_per_second": 5.118,
|
| 1172 |
+
"step": 770
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.4,
|
| 1176 |
+
"grad_norm": 1.5382564067840576,
|
| 1177 |
+
"learning_rate": 1.8829475928589272e-05,
|
| 1178 |
+
"loss": 0.1888,
|
| 1179 |
+
"step": 780
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.4,
|
| 1183 |
+
"eval_loss": 0.8794592022895813,
|
| 1184 |
+
"eval_runtime": 1.1678,
|
| 1185 |
+
"eval_samples_per_second": 97.622,
|
| 1186 |
+
"eval_steps_per_second": 5.138,
|
| 1187 |
+
"step": 780
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.430769230769231,
|
| 1191 |
+
"grad_norm": 2.371277093887329,
|
| 1192 |
+
"learning_rate": 1.8778544091784047e-05,
|
| 1193 |
+
"loss": 0.2293,
|
| 1194 |
+
"step": 790
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.430769230769231,
|
| 1198 |
+
"eval_loss": 0.8893064260482788,
|
| 1199 |
+
"eval_runtime": 1.1684,
|
| 1200 |
+
"eval_samples_per_second": 97.567,
|
| 1201 |
+
"eval_steps_per_second": 5.135,
|
| 1202 |
+
"step": 790
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 2.4615384615384617,
|
| 1206 |
+
"grad_norm": 1.4892629384994507,
|
| 1207 |
+
"learning_rate": 1.8726599588756144e-05,
|
| 1208 |
+
"loss": 0.2326,
|
| 1209 |
+
"step": 800
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 2.4615384615384617,
|
| 1213 |
+
"eval_loss": 0.8676194548606873,
|
| 1214 |
+
"eval_runtime": 1.1699,
|
| 1215 |
+
"eval_samples_per_second": 97.441,
|
| 1216 |
+
"eval_steps_per_second": 5.128,
|
| 1217 |
+
"step": 800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 2.4923076923076923,
|
| 1221 |
+
"grad_norm": 1.6047152280807495,
|
| 1222 |
+
"learning_rate": 1.8673648411665895e-05,
|
| 1223 |
+
"loss": 0.1986,
|
| 1224 |
+
"step": 810
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 2.4923076923076923,
|
| 1228 |
+
"eval_loss": 0.8942193984985352,
|
| 1229 |
+
"eval_runtime": 1.17,
|
| 1230 |
+
"eval_samples_per_second": 97.433,
|
| 1231 |
+
"eval_steps_per_second": 5.128,
|
| 1232 |
+
"step": 810
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 2.523076923076923,
|
| 1236 |
+
"grad_norm": 2.106461524963379,
|
| 1237 |
+
"learning_rate": 1.8619696668800494e-05,
|
| 1238 |
+
"loss": 0.238,
|
| 1239 |
+
"step": 820
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 2.523076923076923,
|
| 1243 |
+
"eval_loss": 0.8907297849655151,
|
| 1244 |
+
"eval_runtime": 1.1681,
|
| 1245 |
+
"eval_samples_per_second": 97.593,
|
| 1246 |
+
"eval_steps_per_second": 5.136,
|
| 1247 |
+
"step": 820
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 2.5538461538461537,
|
| 1251 |
+
"grad_norm": 1.7706941366195679,
|
| 1252 |
+
"learning_rate": 1.8564750583869374e-05,
|
| 1253 |
+
"loss": 0.2278,
|
| 1254 |
+
"step": 830
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 2.5538461538461537,
|
| 1258 |
+
"eval_loss": 0.8588199019432068,
|
| 1259 |
+
"eval_runtime": 1.1709,
|
| 1260 |
+
"eval_samples_per_second": 97.36,
|
| 1261 |
+
"eval_steps_per_second": 5.124,
|
| 1262 |
+
"step": 830
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.5846153846153848,
|
| 1266 |
+
"grad_norm": 1.4446183443069458,
|
| 1267 |
+
"learning_rate": 1.850881649528625e-05,
|
| 1268 |
+
"loss": 0.2033,
|
| 1269 |
+
"step": 840
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.5846153846153848,
|
| 1273 |
+
"eval_loss": 0.8720155954360962,
|
| 1274 |
+
"eval_runtime": 1.1716,
|
| 1275 |
+
"eval_samples_per_second": 97.304,
|
| 1276 |
+
"eval_steps_per_second": 5.121,
|
| 1277 |
+
"step": 840
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.6153846153846154,
|
| 1281 |
+
"grad_norm": 1.5155885219573975,
|
| 1282 |
+
"learning_rate": 1.845190085543795e-05,
|
| 1283 |
+
"loss": 0.2211,
|
| 1284 |
+
"step": 850
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.6153846153846154,
|
| 1288 |
+
"eval_loss": 0.8732415437698364,
|
| 1289 |
+
"eval_runtime": 1.1708,
|
| 1290 |
+
"eval_samples_per_second": 97.371,
|
| 1291 |
+
"eval_steps_per_second": 5.125,
|
| 1292 |
+
"step": 850
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.646153846153846,
|
| 1296 |
+
"grad_norm": 2.934957981109619,
|
| 1297 |
+
"learning_rate": 1.839401022994006e-05,
|
| 1298 |
+
"loss": 0.21,
|
| 1299 |
+
"step": 860
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.646153846153846,
|
| 1303 |
+
"eval_loss": 0.8790932297706604,
|
| 1304 |
+
"eval_runtime": 1.1671,
|
| 1305 |
+
"eval_samples_per_second": 97.68,
|
| 1306 |
+
"eval_steps_per_second": 5.141,
|
| 1307 |
+
"step": 860
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 2.676923076923077,
|
| 1311 |
+
"grad_norm": 2.930457353591919,
|
| 1312 |
+
"learning_rate": 1.8335151296879576e-05,
|
| 1313 |
+
"loss": 0.2313,
|
| 1314 |
+
"step": 870
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 2.676923076923077,
|
| 1318 |
+
"eval_loss": 0.8555310368537903,
|
| 1319 |
+
"eval_runtime": 1.1718,
|
| 1320 |
+
"eval_samples_per_second": 97.284,
|
| 1321 |
+
"eval_steps_per_second": 5.12,
|
| 1322 |
+
"step": 870
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 2.707692307692308,
|
| 1326 |
+
"grad_norm": 2.029303789138794,
|
| 1327 |
+
"learning_rate": 1.82753308460445e-05,
|
| 1328 |
+
"loss": 0.2285,
|
| 1329 |
+
"step": 880
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 2.707692307692308,
|
| 1333 |
+
"eval_loss": 0.870873749256134,
|
| 1334 |
+
"eval_runtime": 1.1696,
|
| 1335 |
+
"eval_samples_per_second": 97.47,
|
| 1336 |
+
"eval_steps_per_second": 5.13,
|
| 1337 |
+
"step": 880
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 2.7384615384615385,
|
| 1341 |
+
"grad_norm": 1.949743390083313,
|
| 1342 |
+
"learning_rate": 1.821455577814062e-05,
|
| 1343 |
+
"loss": 0.2134,
|
| 1344 |
+
"step": 890
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 2.7384615384615385,
|
| 1348 |
+
"eval_loss": 0.8873838186264038,
|
| 1349 |
+
"eval_runtime": 1.1688,
|
| 1350 |
+
"eval_samples_per_second": 97.538,
|
| 1351 |
+
"eval_steps_per_second": 5.134,
|
| 1352 |
+
"step": 890
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 2.769230769230769,
|
| 1356 |
+
"grad_norm": 1.6945637464523315,
|
| 1357 |
+
"learning_rate": 1.8152833103995443e-05,
|
| 1358 |
+
"loss": 0.218,
|
| 1359 |
+
"step": 900
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 2.769230769230769,
|
| 1363 |
+
"eval_loss": 0.8798539042472839,
|
| 1364 |
+
"eval_runtime": 1.1699,
|
| 1365 |
+
"eval_samples_per_second": 97.444,
|
| 1366 |
+
"eval_steps_per_second": 5.129,
|
| 1367 |
+
"step": 900
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8,
|
| 1371 |
+
"grad_norm": 2.2784035205841064,
|
| 1372 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 1373 |
+
"loss": 0.2426,
|
| 1374 |
+
"step": 910
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8,
|
| 1378 |
+
"eval_loss": 0.8753871321678162,
|
| 1379 |
+
"eval_runtime": 1.1723,
|
| 1380 |
+
"eval_samples_per_second": 97.246,
|
| 1381 |
+
"eval_steps_per_second": 5.118,
|
| 1382 |
+
"step": 910
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.830769230769231,
|
| 1386 |
+
"grad_norm": 1.750718593597412,
|
| 1387 |
+
"learning_rate": 1.802657352603483e-05,
|
| 1388 |
+
"loss": 0.2059,
|
| 1389 |
+
"step": 920
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.830769230769231,
|
| 1393 |
+
"eval_loss": 0.8761821985244751,
|
| 1394 |
+
"eval_runtime": 1.1762,
|
| 1395 |
+
"eval_samples_per_second": 96.92,
|
| 1396 |
+
"eval_steps_per_second": 5.101,
|
| 1397 |
+
"step": 920
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.8615384615384616,
|
| 1401 |
+
"grad_norm": 1.2204371690750122,
|
| 1402 |
+
"learning_rate": 1.7962051187141377e-05,
|
| 1403 |
+
"loss": 0.2229,
|
| 1404 |
+
"step": 930
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.8615384615384616,
|
| 1408 |
+
"eval_loss": 0.8772630095481873,
|
| 1409 |
+
"eval_runtime": 1.1697,
|
| 1410 |
+
"eval_samples_per_second": 97.458,
|
| 1411 |
+
"eval_steps_per_second": 5.129,
|
| 1412 |
+
"step": 930
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 2.8923076923076922,
|
| 1416 |
+
"grad_norm": 2.0748281478881836,
|
| 1417 |
+
"learning_rate": 1.7896610370170452e-05,
|
| 1418 |
+
"loss": 0.2582,
|
| 1419 |
+
"step": 940
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 2.8923076923076922,
|
| 1423 |
+
"eval_loss": 0.884286105632782,
|
| 1424 |
+
"eval_runtime": 1.1773,
|
| 1425 |
+
"eval_samples_per_second": 96.829,
|
| 1426 |
+
"eval_steps_per_second": 5.096,
|
| 1427 |
+
"step": 940
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.9230769230769234,
|
| 1431 |
+
"grad_norm": 1.9051027297973633,
|
| 1432 |
+
"learning_rate": 1.7830258624176224e-05,
|
| 1433 |
+
"loss": 0.2102,
|
| 1434 |
+
"step": 950
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.9230769230769234,
|
| 1438 |
+
"eval_loss": 0.878767192363739,
|
| 1439 |
+
"eval_runtime": 1.1684,
|
| 1440 |
+
"eval_samples_per_second": 97.571,
|
| 1441 |
+
"eval_steps_per_second": 5.135,
|
| 1442 |
+
"step": 950
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 2.953846153846154,
|
| 1446 |
+
"grad_norm": 1.8921849727630615,
|
| 1447 |
+
"learning_rate": 1.776300360329488e-05,
|
| 1448 |
+
"loss": 0.2331,
|
| 1449 |
+
"step": 960
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 2.953846153846154,
|
| 1453 |
+
"eval_loss": 0.8681402206420898,
|
| 1454 |
+
"eval_runtime": 1.1693,
|
| 1455 |
+
"eval_samples_per_second": 97.497,
|
| 1456 |
+
"eval_steps_per_second": 5.131,
|
| 1457 |
+
"step": 960
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.9846153846153847,
|
| 1461 |
+
"grad_norm": 1.5593761205673218,
|
| 1462 |
+
"learning_rate": 1.769485306586166e-05,
|
| 1463 |
+
"loss": 0.2226,
|
| 1464 |
+
"step": 970
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 2.9846153846153847,
|
| 1468 |
+
"eval_loss": 0.8726215958595276,
|
| 1469 |
+
"eval_runtime": 1.1712,
|
| 1470 |
+
"eval_samples_per_second": 97.336,
|
| 1471 |
+
"eval_steps_per_second": 5.123,
|
| 1472 |
+
"step": 970
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 3.0153846153846153,
|
| 1476 |
+
"grad_norm": 1.4075913429260254,
|
| 1477 |
+
"learning_rate": 1.762581487351587e-05,
|
| 1478 |
+
"loss": 0.1535,
|
| 1479 |
+
"step": 980
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 3.0153846153846153,
|
| 1483 |
+
"eval_loss": 0.9113463163375854,
|
| 1484 |
+
"eval_runtime": 1.1794,
|
| 1485 |
+
"eval_samples_per_second": 96.661,
|
| 1486 |
+
"eval_steps_per_second": 5.087,
|
| 1487 |
+
"step": 980
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 3.046153846153846,
|
| 1491 |
+
"grad_norm": 1.410906434059143,
|
| 1492 |
+
"learning_rate": 1.7555896990294003e-05,
|
| 1493 |
+
"loss": 0.1021,
|
| 1494 |
+
"step": 990
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 3.046153846153846,
|
| 1498 |
+
"eval_loss": 0.9633604288101196,
|
| 1499 |
+
"eval_runtime": 1.1702,
|
| 1500 |
+
"eval_samples_per_second": 97.416,
|
| 1501 |
+
"eval_steps_per_second": 5.127,
|
| 1502 |
+
"step": 990
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 3.076923076923077,
|
| 1506 |
+
"grad_norm": 1.0470328330993652,
|
| 1507 |
+
"learning_rate": 1.7485107481711014e-05,
|
| 1508 |
+
"loss": 0.1118,
|
| 1509 |
+
"step": 1000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 3.076923076923077,
|
| 1513 |
+
"eval_loss": 0.9647616744041443,
|
| 1514 |
+
"eval_runtime": 1.1683,
|
| 1515 |
+
"eval_samples_per_second": 97.574,
|
| 1516 |
+
"eval_steps_per_second": 5.135,
|
| 1517 |
+
"step": 1000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 3.1076923076923078,
|
| 1521 |
+
"grad_norm": 2.098592758178711,
|
| 1522 |
+
"learning_rate": 1.741345451382992e-05,
|
| 1523 |
+
"loss": 0.1224,
|
| 1524 |
+
"step": 1010
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 3.1076923076923078,
|
| 1528 |
+
"eval_loss": 0.9478728175163269,
|
| 1529 |
+
"eval_runtime": 1.1735,
|
| 1530 |
+
"eval_samples_per_second": 97.143,
|
| 1531 |
+
"eval_steps_per_second": 5.113,
|
| 1532 |
+
"step": 1010
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 3.1384615384615384,
|
| 1536 |
+
"grad_norm": 1.6263700723648071,
|
| 1537 |
+
"learning_rate": 1.7340946352319795e-05,
|
| 1538 |
+
"loss": 0.1377,
|
| 1539 |
+
"step": 1020
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 3.1384615384615384,
|
| 1543 |
+
"eval_loss": 0.922594428062439,
|
| 1544 |
+
"eval_runtime": 1.1721,
|
| 1545 |
+
"eval_samples_per_second": 97.262,
|
| 1546 |
+
"eval_steps_per_second": 5.119,
|
| 1547 |
+
"step": 1020
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 3.169230769230769,
|
| 1551 |
+
"grad_norm": 1.0413360595703125,
|
| 1552 |
+
"learning_rate": 1.7267591361502233e-05,
|
| 1553 |
+
"loss": 0.1249,
|
| 1554 |
+
"step": 1030
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 3.169230769230769,
|
| 1558 |
+
"eval_loss": 0.9370183944702148,
|
| 1559 |
+
"eval_runtime": 1.169,
|
| 1560 |
+
"eval_samples_per_second": 97.519,
|
| 1561 |
+
"eval_steps_per_second": 5.133,
|
| 1562 |
+
"step": 1030
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 3.2,
|
| 1566 |
+
"grad_norm": 1.6745877265930176,
|
| 1567 |
+
"learning_rate": 1.7193398003386514e-05,
|
| 1568 |
+
"loss": 0.1186,
|
| 1569 |
+
"step": 1040
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 3.2,
|
| 1573 |
+
"eval_loss": 0.9573487639427185,
|
| 1574 |
+
"eval_runtime": 1.1681,
|
| 1575 |
+
"eval_samples_per_second": 97.596,
|
| 1576 |
+
"eval_steps_per_second": 5.137,
|
| 1577 |
+
"step": 1040
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 3.230769230769231,
|
| 1581 |
+
"grad_norm": 1.3979861736297607,
|
| 1582 |
+
"learning_rate": 1.7118374836693407e-05,
|
| 1583 |
+
"loss": 0.135,
|
| 1584 |
+
"step": 1050
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 3.230769230769231,
|
| 1588 |
+
"eval_loss": 0.9554935693740845,
|
| 1589 |
+
"eval_runtime": 1.1664,
|
| 1590 |
+
"eval_samples_per_second": 97.735,
|
| 1591 |
+
"eval_steps_per_second": 5.144,
|
| 1592 |
+
"step": 1050
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 3.2615384615384615,
|
| 1596 |
+
"grad_norm": 1.4017126560211182,
|
| 1597 |
+
"learning_rate": 1.7042530515867897e-05,
|
| 1598 |
+
"loss": 0.1102,
|
| 1599 |
+
"step": 1060
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 3.2615384615384615,
|
| 1603 |
+
"eval_loss": 0.9547311663627625,
|
| 1604 |
+
"eval_runtime": 1.1685,
|
| 1605 |
+
"eval_samples_per_second": 97.561,
|
| 1606 |
+
"eval_steps_per_second": 5.135,
|
| 1607 |
+
"step": 1060
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 3.292307692307692,
|
| 1611 |
+
"grad_norm": 1.3390568494796753,
|
| 1612 |
+
"learning_rate": 1.6965873790080806e-05,
|
| 1613 |
+
"loss": 0.1358,
|
| 1614 |
+
"step": 1070
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 3.292307692307692,
|
| 1618 |
+
"eval_loss": 0.9695214629173279,
|
| 1619 |
+
"eval_runtime": 1.1695,
|
| 1620 |
+
"eval_samples_per_second": 97.474,
|
| 1621 |
+
"eval_steps_per_second": 5.13,
|
| 1622 |
+
"step": 1070
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 3.3230769230769233,
|
| 1626 |
+
"grad_norm": 1.9002848863601685,
|
| 1627 |
+
"learning_rate": 1.6888413502219534e-05,
|
| 1628 |
+
"loss": 0.1419,
|
| 1629 |
+
"step": 1080
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 3.3230769230769233,
|
| 1633 |
+
"eval_loss": 0.9528248310089111,
|
| 1634 |
+
"eval_runtime": 1.174,
|
| 1635 |
+
"eval_samples_per_second": 97.107,
|
| 1636 |
+
"eval_steps_per_second": 5.111,
|
| 1637 |
+
"step": 1080
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 3.353846153846154,
|
| 1641 |
+
"grad_norm": 1.7857142686843872,
|
| 1642 |
+
"learning_rate": 1.6810158587867973e-05,
|
| 1643 |
+
"loss": 0.128,
|
| 1644 |
+
"step": 1090
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 3.353846153846154,
|
| 1648 |
+
"eval_loss": 0.9331367611885071,
|
| 1649 |
+
"eval_runtime": 1.1718,
|
| 1650 |
+
"eval_samples_per_second": 97.282,
|
| 1651 |
+
"eval_steps_per_second": 5.12,
|
| 1652 |
+
"step": 1090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 3.3846153846153846,
|
| 1656 |
+
"grad_norm": 1.5178117752075195,
|
| 1657 |
+
"learning_rate": 1.67311180742757e-05,
|
| 1658 |
+
"loss": 0.112,
|
| 1659 |
+
"step": 1100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 3.3846153846153846,
|
| 1663 |
+
"eval_loss": 0.9547919631004333,
|
| 1664 |
+
"eval_runtime": 1.1695,
|
| 1665 |
+
"eval_samples_per_second": 97.48,
|
| 1666 |
+
"eval_steps_per_second": 5.131,
|
| 1667 |
+
"step": 1100
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 3.4153846153846152,
|
| 1671 |
+
"grad_norm": 1.3151490688323975,
|
| 1672 |
+
"learning_rate": 1.665130107931666e-05,
|
| 1673 |
+
"loss": 0.1119,
|
| 1674 |
+
"step": 1110
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 3.4153846153846152,
|
| 1678 |
+
"eval_loss": 0.9812674522399902,
|
| 1679 |
+
"eval_runtime": 1.1666,
|
| 1680 |
+
"eval_samples_per_second": 97.721,
|
| 1681 |
+
"eval_steps_per_second": 5.143,
|
| 1682 |
+
"step": 1110
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 3.4461538461538463,
|
| 1686 |
+
"grad_norm": 1.4334938526153564,
|
| 1687 |
+
"learning_rate": 1.657071681043731e-05,
|
| 1688 |
+
"loss": 0.1042,
|
| 1689 |
+
"step": 1120
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 3.4461538461538463,
|
| 1693 |
+
"eval_loss": 0.9601649641990662,
|
| 1694 |
+
"eval_runtime": 1.1706,
|
| 1695 |
+
"eval_samples_per_second": 97.388,
|
| 1696 |
+
"eval_steps_per_second": 5.126,
|
| 1697 |
+
"step": 1120
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 3.476923076923077,
|
| 1701 |
+
"grad_norm": 1.3328880071640015,
|
| 1702 |
+
"learning_rate": 1.648937456359451e-05,
|
| 1703 |
+
"loss": 0.1129,
|
| 1704 |
+
"step": 1130
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 3.476923076923077,
|
| 1708 |
+
"eval_loss": 0.9889415502548218,
|
| 1709 |
+
"eval_runtime": 1.1689,
|
| 1710 |
+
"eval_samples_per_second": 97.527,
|
| 1711 |
+
"eval_steps_per_second": 5.133,
|
| 1712 |
+
"step": 1130
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 3.5076923076923077,
|
| 1716 |
+
"grad_norm": 1.8561350107192993,
|
| 1717 |
+
"learning_rate": 1.640728372218317e-05,
|
| 1718 |
+
"loss": 0.1247,
|
| 1719 |
+
"step": 1140
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 3.5076923076923077,
|
| 1723 |
+
"eval_loss": 0.9682590961456299,
|
| 1724 |
+
"eval_runtime": 1.1697,
|
| 1725 |
+
"eval_samples_per_second": 97.461,
|
| 1726 |
+
"eval_steps_per_second": 5.13,
|
| 1727 |
+
"step": 1140
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 3.5384615384615383,
|
| 1731 |
+
"grad_norm": 1.1813424825668335,
|
| 1732 |
+
"learning_rate": 1.6324453755953772e-05,
|
| 1733 |
+
"loss": 0.1213,
|
| 1734 |
+
"step": 1150
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 3.5384615384615383,
|
| 1738 |
+
"eval_loss": 0.9461174607276917,
|
| 1739 |
+
"eval_runtime": 1.1671,
|
| 1740 |
+
"eval_samples_per_second": 97.674,
|
| 1741 |
+
"eval_steps_per_second": 5.141,
|
| 1742 |
+
"step": 1150
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 3.569230769230769,
|
| 1746 |
+
"grad_norm": 1.2129303216934204,
|
| 1747 |
+
"learning_rate": 1.624089421992003e-05,
|
| 1748 |
+
"loss": 0.1253,
|
| 1749 |
+
"step": 1160
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 3.569230769230769,
|
| 1753 |
+
"eval_loss": 0.9655346274375916,
|
| 1754 |
+
"eval_runtime": 1.1837,
|
| 1755 |
+
"eval_samples_per_second": 96.312,
|
| 1756 |
+
"eval_steps_per_second": 5.069,
|
| 1757 |
+
"step": 1160
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 3.6,
|
| 1761 |
+
"grad_norm": 2.2233059406280518,
|
| 1762 |
+
"learning_rate": 1.6156614753256583e-05,
|
| 1763 |
+
"loss": 0.1147,
|
| 1764 |
+
"step": 1170
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 3.6,
|
| 1768 |
+
"eval_loss": 0.9837222099304199,
|
| 1769 |
+
"eval_runtime": 1.168,
|
| 1770 |
+
"eval_samples_per_second": 97.6,
|
| 1771 |
+
"eval_steps_per_second": 5.137,
|
| 1772 |
+
"step": 1170
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 3.6307692307692307,
|
| 1776 |
+
"grad_norm": 1.382871389389038,
|
| 1777 |
+
"learning_rate": 1.6071625078187113e-05,
|
| 1778 |
+
"loss": 0.1272,
|
| 1779 |
+
"step": 1180
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 3.6307692307692307,
|
| 1783 |
+
"eval_loss": 0.9973061680793762,
|
| 1784 |
+
"eval_runtime": 1.1757,
|
| 1785 |
+
"eval_samples_per_second": 96.964,
|
| 1786 |
+
"eval_steps_per_second": 5.103,
|
| 1787 |
+
"step": 1180
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 3.6615384615384614,
|
| 1791 |
+
"grad_norm": 1.2613117694854736,
|
| 1792 |
+
"learning_rate": 1.5985934998862775e-05,
|
| 1793 |
+
"loss": 0.1367,
|
| 1794 |
+
"step": 1190
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 3.6615384615384614,
|
| 1798 |
+
"eval_loss": 0.9685854315757751,
|
| 1799 |
+
"eval_runtime": 1.1695,
|
| 1800 |
+
"eval_samples_per_second": 97.474,
|
| 1801 |
+
"eval_steps_per_second": 5.13,
|
| 1802 |
+
"step": 1190
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 3.6923076923076925,
|
| 1806 |
+
"grad_norm": 1.0769059658050537,
|
| 1807 |
+
"learning_rate": 1.5899554400231233e-05,
|
| 1808 |
+
"loss": 0.1208,
|
| 1809 |
+
"step": 1200
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 3.6923076923076925,
|
| 1813 |
+
"eval_loss": 0.9511159062385559,
|
| 1814 |
+
"eval_runtime": 1.1684,
|
| 1815 |
+
"eval_samples_per_second": 97.566,
|
| 1816 |
+
"eval_steps_per_second": 5.135,
|
| 1817 |
+
"step": 1200
|
| 1818 |
+
}
|
| 1819 |
+
],
|
| 1820 |
+
"logging_steps": 10,
|
| 1821 |
+
"max_steps": 3250,
|
| 1822 |
+
"num_input_tokens_seen": 0,
|
| 1823 |
+
"num_train_epochs": 10,
|
| 1824 |
+
"save_steps": 50,
|
| 1825 |
+
"total_flos": 7.490852910582989e+16,
|
| 1826 |
+
"train_batch_size": 2,
|
| 1827 |
+
"trial_name": null,
|
| 1828 |
+
"trial_params": null
|
| 1829 |
+
}
|