End of training
Browse files- README.md +3 -3
- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +10 -0
- trainer_state.json +485 -0
README.md
CHANGED
|
@@ -18,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [EleutherAI/pythia-410m](https://huggingface.co/EleutherAI/pythia-410m) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 1.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
-
- Num Input Tokens Seen:
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [EleutherAI/pythia-410m](https://huggingface.co/EleutherAI/pythia-410m) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.6663
|
| 22 |
+
- Accuracy: 0.1806
|
| 23 |
+
- Num Input Tokens Seen: 1990656
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.18055555555555555,
|
| 4 |
+
"eval_loss": 1.6663275957107544,
|
| 5 |
+
"eval_runtime": 0.372,
|
| 6 |
+
"eval_samples": 72,
|
| 7 |
+
"eval_samples_per_second": 193.541,
|
| 8 |
+
"eval_steps_per_second": 24.193,
|
| 9 |
+
"num_input_tokens_seen": 1990656,
|
| 10 |
+
"total_flos": 3610849453277184.0,
|
| 11 |
+
"train_loss": 1.8222727520475663,
|
| 12 |
+
"train_runtime": 109.7362,
|
| 13 |
+
"train_samples": 645,
|
| 14 |
+
"train_samples_per_second": 17.633,
|
| 15 |
+
"train_steps_per_second": 2.214
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.18055555555555555,
|
| 4 |
+
"eval_loss": 1.6663275957107544,
|
| 5 |
+
"eval_runtime": 0.372,
|
| 6 |
+
"eval_samples": 72,
|
| 7 |
+
"eval_samples_per_second": 193.541,
|
| 8 |
+
"eval_steps_per_second": 24.193,
|
| 9 |
+
"num_input_tokens_seen": 1990656
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"num_input_tokens_seen": 1990656,
|
| 4 |
+
"total_flos": 3610849453277184.0,
|
| 5 |
+
"train_loss": 1.8222727520475663,
|
| 6 |
+
"train_runtime": 109.7362,
|
| 7 |
+
"train_samples": 645,
|
| 8 |
+
"train_samples_per_second": 17.633,
|
| 9 |
+
"train_steps_per_second": 2.214
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,485 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.6663275957107544,
|
| 3 |
+
"best_model_checkpoint": "/data/user_data/gonilude/java_and_text_pythia_410m/checkpoint-150",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 243,
|
| 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,
|
| 13 |
+
"eval_accuracy": 0.25,
|
| 14 |
+
"eval_loss": 2.6309967041015625,
|
| 15 |
+
"eval_runtime": 0.5335,
|
| 16 |
+
"eval_samples_per_second": 134.957,
|
| 17 |
+
"eval_steps_per_second": 16.87,
|
| 18 |
+
"num_input_tokens_seen": 0,
|
| 19 |
+
"step": 0
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 0.012345679012345678,
|
| 23 |
+
"grad_norm": NaN,
|
| 24 |
+
"learning_rate": 0.0,
|
| 25 |
+
"loss": 3.7346,
|
| 26 |
+
"num_input_tokens_seen": 8192,
|
| 27 |
+
"step": 1
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.06172839506172839,
|
| 31 |
+
"grad_norm": Infinity,
|
| 32 |
+
"learning_rate": 0.0,
|
| 33 |
+
"loss": 3.0853,
|
| 34 |
+
"num_input_tokens_seen": 40960,
|
| 35 |
+
"step": 5
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.12345679012345678,
|
| 39 |
+
"grad_norm": 140.6628875732422,
|
| 40 |
+
"learning_rate": 1e-05,
|
| 41 |
+
"loss": 2.951,
|
| 42 |
+
"num_input_tokens_seen": 81920,
|
| 43 |
+
"step": 10
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.18518518518518517,
|
| 47 |
+
"grad_norm": 121.90998840332031,
|
| 48 |
+
"learning_rate": 1.9999106432103785e-05,
|
| 49 |
+
"loss": 3.6623,
|
| 50 |
+
"num_input_tokens_seen": 122880,
|
| 51 |
+
"step": 15
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.24691358024691357,
|
| 55 |
+
"grad_norm": 119.89352416992188,
|
| 56 |
+
"learning_rate": 1.9967848320275253e-05,
|
| 57 |
+
"loss": 2.5551,
|
| 58 |
+
"num_input_tokens_seen": 163840,
|
| 59 |
+
"step": 20
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.30864197530864196,
|
| 63 |
+
"grad_norm": 69.07218170166016,
|
| 64 |
+
"learning_rate": 1.9892071378116378e-05,
|
| 65 |
+
"loss": 2.146,
|
| 66 |
+
"num_input_tokens_seen": 204800,
|
| 67 |
+
"step": 25
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.37037037037037035,
|
| 71 |
+
"grad_norm": 38.68893051147461,
|
| 72 |
+
"learning_rate": 1.9772114043845968e-05,
|
| 73 |
+
"loss": 2.4626,
|
| 74 |
+
"num_input_tokens_seen": 245760,
|
| 75 |
+
"step": 30
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.43209876543209874,
|
| 79 |
+
"grad_norm": 12.328490257263184,
|
| 80 |
+
"learning_rate": 1.9608512076038964e-05,
|
| 81 |
+
"loss": 1.976,
|
| 82 |
+
"num_input_tokens_seen": 286720,
|
| 83 |
+
"step": 35
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.49382716049382713,
|
| 87 |
+
"grad_norm": 28.689950942993164,
|
| 88 |
+
"learning_rate": 1.9401996160798574e-05,
|
| 89 |
+
"loss": 1.8516,
|
| 90 |
+
"num_input_tokens_seen": 327680,
|
| 91 |
+
"step": 40
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.5555555555555556,
|
| 95 |
+
"grad_norm": 20.923078536987305,
|
| 96 |
+
"learning_rate": 1.915348864833476e-05,
|
| 97 |
+
"loss": 1.669,
|
| 98 |
+
"num_input_tokens_seen": 368640,
|
| 99 |
+
"step": 45
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.6172839506172839,
|
| 103 |
+
"grad_norm": 60.305381774902344,
|
| 104 |
+
"learning_rate": 1.8864099433524302e-05,
|
| 105 |
+
"loss": 1.9603,
|
| 106 |
+
"num_input_tokens_seen": 409600,
|
| 107 |
+
"step": 50
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6172839506172839,
|
| 111 |
+
"eval_accuracy": 0.19444444444444445,
|
| 112 |
+
"eval_loss": 1.8224691152572632,
|
| 113 |
+
"eval_runtime": 0.3688,
|
| 114 |
+
"eval_samples_per_second": 195.251,
|
| 115 |
+
"eval_steps_per_second": 24.406,
|
| 116 |
+
"num_input_tokens_seen": 409600,
|
| 117 |
+
"step": 50
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.6790123456790124,
|
| 121 |
+
"grad_norm": 37.31528854370117,
|
| 122 |
+
"learning_rate": 1.853512099885085e-05,
|
| 123 |
+
"loss": 1.6856,
|
| 124 |
+
"num_input_tokens_seen": 450560,
|
| 125 |
+
"step": 55
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.7407407407407407,
|
| 129 |
+
"grad_norm": 34.03938674926758,
|
| 130 |
+
"learning_rate": 1.816802264186438e-05,
|
| 131 |
+
"loss": 1.989,
|
| 132 |
+
"num_input_tokens_seen": 491520,
|
| 133 |
+
"step": 60
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.8024691358024691,
|
| 137 |
+
"grad_norm": 27.516170501708984,
|
| 138 |
+
"learning_rate": 1.7764443912941675e-05,
|
| 139 |
+
"loss": 1.8994,
|
| 140 |
+
"num_input_tokens_seen": 532480,
|
| 141 |
+
"step": 65
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.8641975308641975,
|
| 145 |
+
"grad_norm": 23.523881912231445,
|
| 146 |
+
"learning_rate": 1.7326187292656332e-05,
|
| 147 |
+
"loss": 1.7613,
|
| 148 |
+
"num_input_tokens_seen": 573440,
|
| 149 |
+
"step": 70
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.9259259259259259,
|
| 153 |
+
"grad_norm": 25.84837532043457,
|
| 154 |
+
"learning_rate": 1.6855210141462964e-05,
|
| 155 |
+
"loss": 1.9531,
|
| 156 |
+
"num_input_tokens_seen": 614400,
|
| 157 |
+
"step": 75
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.9876543209876543,
|
| 161 |
+
"grad_norm": 34.3070068359375,
|
| 162 |
+
"learning_rate": 1.635361595765024e-05,
|
| 163 |
+
"loss": 1.9051,
|
| 164 |
+
"num_input_tokens_seen": 655360,
|
| 165 |
+
"step": 80
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.0493827160493827,
|
| 169 |
+
"grad_norm": 31.36409568786621,
|
| 170 |
+
"learning_rate": 1.5823644982606905e-05,
|
| 171 |
+
"loss": 1.8392,
|
| 172 |
+
"num_input_tokens_seen": 696320,
|
| 173 |
+
"step": 85
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.1111111111111112,
|
| 177 |
+
"grad_norm": 27.246265411376953,
|
| 178 |
+
"learning_rate": 1.5267664195359917e-05,
|
| 179 |
+
"loss": 1.6923,
|
| 180 |
+
"num_input_tokens_seen": 737280,
|
| 181 |
+
"step": 90
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.1728395061728394,
|
| 185 |
+
"grad_norm": 41.13190841674805,
|
| 186 |
+
"learning_rate": 1.4688156741071513e-05,
|
| 187 |
+
"loss": 1.8443,
|
| 188 |
+
"num_input_tokens_seen": 778240,
|
| 189 |
+
"step": 95
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.2345679012345678,
|
| 193 |
+
"grad_norm": 38.65595626831055,
|
| 194 |
+
"learning_rate": 1.408771084071012e-05,
|
| 195 |
+
"loss": 1.7774,
|
| 196 |
+
"num_input_tokens_seen": 819200,
|
| 197 |
+
"step": 100
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.2345679012345678,
|
| 201 |
+
"eval_accuracy": 0.18055555555555555,
|
| 202 |
+
"eval_loss": 1.7543673515319824,
|
| 203 |
+
"eval_runtime": 0.371,
|
| 204 |
+
"eval_samples_per_second": 194.075,
|
| 205 |
+
"eval_steps_per_second": 24.259,
|
| 206 |
+
"num_input_tokens_seen": 819200,
|
| 207 |
+
"step": 100
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.2962962962962963,
|
| 211 |
+
"grad_norm": 14.981557846069336,
|
| 212 |
+
"learning_rate": 1.3469008231427207e-05,
|
| 213 |
+
"loss": 1.8287,
|
| 214 |
+
"num_input_tokens_seen": 860160,
|
| 215 |
+
"step": 105
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.3580246913580247,
|
| 219 |
+
"grad_norm": 21.5086727142334,
|
| 220 |
+
"learning_rate": 1.283481218926818e-05,
|
| 221 |
+
"loss": 1.5601,
|
| 222 |
+
"num_input_tokens_seen": 901120,
|
| 223 |
+
"step": 110
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.4197530864197532,
|
| 227 |
+
"grad_norm": 24.150880813598633,
|
| 228 |
+
"learning_rate": 1.2187955187710752e-05,
|
| 229 |
+
"loss": 1.8594,
|
| 230 |
+
"num_input_tokens_seen": 942080,
|
| 231 |
+
"step": 115
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.4814814814814814,
|
| 235 |
+
"grad_norm": 11.27273941040039,
|
| 236 |
+
"learning_rate": 1.1531326247150802e-05,
|
| 237 |
+
"loss": 1.6694,
|
| 238 |
+
"num_input_tokens_seen": 983040,
|
| 239 |
+
"step": 120
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.5432098765432098,
|
| 243 |
+
"grad_norm": 24.10595703125,
|
| 244 |
+
"learning_rate": 1.0867858031835975e-05,
|
| 245 |
+
"loss": 1.6684,
|
| 246 |
+
"num_input_tokens_seen": 1024000,
|
| 247 |
+
"step": 125
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.6049382716049383,
|
| 251 |
+
"grad_norm": 15.398138046264648,
|
| 252 |
+
"learning_rate": 1.0200513751875227e-05,
|
| 253 |
+
"loss": 1.6126,
|
| 254 |
+
"num_input_tokens_seen": 1064960,
|
| 255 |
+
"step": 130
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.6666666666666665,
|
| 259 |
+
"grad_norm": 25.667972564697266,
|
| 260 |
+
"learning_rate": 9.532273928823151e-06,
|
| 261 |
+
"loss": 1.5059,
|
| 262 |
+
"num_input_tokens_seen": 1105920,
|
| 263 |
+
"step": 135
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.7283950617283952,
|
| 267 |
+
"grad_norm": 21.25739288330078,
|
| 268 |
+
"learning_rate": 8.866123083947182e-06,
|
| 269 |
+
"loss": 1.6684,
|
| 270 |
+
"num_input_tokens_seen": 1146880,
|
| 271 |
+
"step": 140
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 1.7901234567901234,
|
| 275 |
+
"grad_norm": 17.899890899658203,
|
| 276 |
+
"learning_rate": 8.20503640863111e-06,
|
| 277 |
+
"loss": 1.5919,
|
| 278 |
+
"num_input_tokens_seen": 1187840,
|
| 279 |
+
"step": 145
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.8518518518518519,
|
| 283 |
+
"grad_norm": 25.205127716064453,
|
| 284 |
+
"learning_rate": 7.55196647644814e-06,
|
| 285 |
+
"loss": 1.6193,
|
| 286 |
+
"num_input_tokens_seen": 1228800,
|
| 287 |
+
"step": 150
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.8518518518518519,
|
| 291 |
+
"eval_accuracy": 0.18055555555555555,
|
| 292 |
+
"eval_loss": 1.6663275957107544,
|
| 293 |
+
"eval_runtime": 0.3724,
|
| 294 |
+
"eval_samples_per_second": 193.328,
|
| 295 |
+
"eval_steps_per_second": 24.166,
|
| 296 |
+
"num_input_tokens_seen": 1228800,
|
| 297 |
+
"step": 150
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.9135802469135803,
|
| 301 |
+
"grad_norm": 32.01460647583008,
|
| 302 |
+
"learning_rate": 6.909830056250527e-06,
|
| 303 |
+
"loss": 1.649,
|
| 304 |
+
"num_input_tokens_seen": 1269760,
|
| 305 |
+
"step": 155
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.9753086419753085,
|
| 309 |
+
"grad_norm": 23.55994415283203,
|
| 310 |
+
"learning_rate": 6.2814950851718695e-06,
|
| 311 |
+
"loss": 1.7344,
|
| 312 |
+
"num_input_tokens_seen": 1310720,
|
| 313 |
+
"step": 160
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.037037037037037,
|
| 317 |
+
"grad_norm": 40.77457046508789,
|
| 318 |
+
"learning_rate": 5.669767859723636e-06,
|
| 319 |
+
"loss": 1.5739,
|
| 320 |
+
"num_input_tokens_seen": 1351680,
|
| 321 |
+
"step": 165
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.0987654320987654,
|
| 325 |
+
"grad_norm": 23.66048240661621,
|
| 326 |
+
"learning_rate": 5.077380502193725e-06,
|
| 327 |
+
"loss": 1.4769,
|
| 328 |
+
"num_input_tokens_seen": 1392640,
|
| 329 |
+
"step": 170
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.1604938271604937,
|
| 333 |
+
"grad_norm": 20.150760650634766,
|
| 334 |
+
"learning_rate": 4.5069787583250815e-06,
|
| 335 |
+
"loss": 1.5486,
|
| 336 |
+
"num_input_tokens_seen": 1433600,
|
| 337 |
+
"step": 175
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 2.2222222222222223,
|
| 341 |
+
"grad_norm": 11.237411499023438,
|
| 342 |
+
"learning_rate": 3.961110180772955e-06,
|
| 343 |
+
"loss": 1.5313,
|
| 344 |
+
"num_input_tokens_seen": 1474560,
|
| 345 |
+
"step": 180
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.2839506172839505,
|
| 349 |
+
"grad_norm": 30.582534790039062,
|
| 350 |
+
"learning_rate": 3.442212751116305e-06,
|
| 351 |
+
"loss": 1.6429,
|
| 352 |
+
"num_input_tokens_seen": 1515520,
|
| 353 |
+
"step": 185
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.3456790123456788,
|
| 357 |
+
"grad_norm": 24.695331573486328,
|
| 358 |
+
"learning_rate": 2.9526039912402504e-06,
|
| 359 |
+
"loss": 1.4812,
|
| 360 |
+
"num_input_tokens_seen": 1556480,
|
| 361 |
+
"step": 190
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.4074074074074074,
|
| 365 |
+
"grad_norm": 6.564646244049072,
|
| 366 |
+
"learning_rate": 2.4944706127207252e-06,
|
| 367 |
+
"loss": 1.5954,
|
| 368 |
+
"num_input_tokens_seen": 1597440,
|
| 369 |
+
"step": 195
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.4691358024691357,
|
| 373 |
+
"grad_norm": 31.85268783569336,
|
| 374 |
+
"learning_rate": 2.0698587504397684e-06,
|
| 375 |
+
"loss": 1.7311,
|
| 376 |
+
"num_input_tokens_seen": 1638400,
|
| 377 |
+
"step": 200
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.4691358024691357,
|
| 381 |
+
"eval_accuracy": 0.19444444444444445,
|
| 382 |
+
"eval_loss": 1.7398681640625,
|
| 383 |
+
"eval_runtime": 0.4238,
|
| 384 |
+
"eval_samples_per_second": 169.875,
|
| 385 |
+
"eval_steps_per_second": 21.234,
|
| 386 |
+
"num_input_tokens_seen": 1638400,
|
| 387 |
+
"step": 200
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.5308641975308643,
|
| 391 |
+
"grad_norm": 29.221206665039062,
|
| 392 |
+
"learning_rate": 1.680664824050432e-06,
|
| 393 |
+
"loss": 1.4484,
|
| 394 |
+
"num_input_tokens_seen": 1679360,
|
| 395 |
+
"step": 205
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.5925925925925926,
|
| 399 |
+
"grad_norm": 35.87575149536133,
|
| 400 |
+
"learning_rate": 1.3286270681062275e-06,
|
| 401 |
+
"loss": 1.5614,
|
| 402 |
+
"num_input_tokens_seen": 1720320,
|
| 403 |
+
"step": 210
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.6543209876543212,
|
| 407 |
+
"grad_norm": 29.08688735961914,
|
| 408 |
+
"learning_rate": 1.015317768683669e-06,
|
| 409 |
+
"loss": 1.5382,
|
| 410 |
+
"num_input_tokens_seen": 1761280,
|
| 411 |
+
"step": 215
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.7160493827160495,
|
| 415 |
+
"grad_norm": 23.70969581604004,
|
| 416 |
+
"learning_rate": 7.421362411709676e-07,
|
| 417 |
+
"loss": 1.5272,
|
| 418 |
+
"num_input_tokens_seen": 1802240,
|
| 419 |
+
"step": 220
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.7777777777777777,
|
| 423 |
+
"grad_norm": 35.782222747802734,
|
| 424 |
+
"learning_rate": 5.103025805858197e-07,
|
| 425 |
+
"loss": 1.5842,
|
| 426 |
+
"num_input_tokens_seen": 1843200,
|
| 427 |
+
"step": 225
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 2.8395061728395063,
|
| 431 |
+
"grad_norm": 31.512516021728516,
|
| 432 |
+
"learning_rate": 3.2085221233487564e-07,
|
| 433 |
+
"loss": 1.5197,
|
| 434 |
+
"num_input_tokens_seen": 1884160,
|
| 435 |
+
"step": 230
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.9012345679012346,
|
| 439 |
+
"grad_norm": 24.439847946166992,
|
| 440 |
+
"learning_rate": 1.7463126775252192e-07,
|
| 441 |
+
"loss": 1.5678,
|
| 442 |
+
"num_input_tokens_seen": 1925120,
|
| 443 |
+
"step": 235
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.962962962962963,
|
| 447 |
+
"grad_norm": 28.297935485839844,
|
| 448 |
+
"learning_rate": 7.229280507293657e-08,
|
| 449 |
+
"loss": 1.6146,
|
| 450 |
+
"num_input_tokens_seen": 1966080,
|
| 451 |
+
"step": 240
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 3.0,
|
| 455 |
+
"num_input_tokens_seen": 1990656,
|
| 456 |
+
"step": 243,
|
| 457 |
+
"total_flos": 3610849453277184.0,
|
| 458 |
+
"train_loss": 1.8222727520475663,
|
| 459 |
+
"train_runtime": 109.7362,
|
| 460 |
+
"train_samples_per_second": 17.633,
|
| 461 |
+
"train_steps_per_second": 2.214
|
| 462 |
+
}
|
| 463 |
+
],
|
| 464 |
+
"logging_steps": 5,
|
| 465 |
+
"max_steps": 243,
|
| 466 |
+
"num_input_tokens_seen": 1990656,
|
| 467 |
+
"num_train_epochs": 3,
|
| 468 |
+
"save_steps": 50,
|
| 469 |
+
"stateful_callbacks": {
|
| 470 |
+
"TrainerControl": {
|
| 471 |
+
"args": {
|
| 472 |
+
"should_epoch_stop": false,
|
| 473 |
+
"should_evaluate": false,
|
| 474 |
+
"should_log": false,
|
| 475 |
+
"should_save": true,
|
| 476 |
+
"should_training_stop": true
|
| 477 |
+
},
|
| 478 |
+
"attributes": {}
|
| 479 |
+
}
|
| 480 |
+
},
|
| 481 |
+
"total_flos": 3610849453277184.0,
|
| 482 |
+
"train_batch_size": 4,
|
| 483 |
+
"trial_name": null,
|
| 484 |
+
"trial_params": null
|
| 485 |
+
}
|