End of training
Browse files- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +136 -0
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"train_loss": 1.6477281238721764,
|
| 4 |
+
"train_runtime": 4221.2735,
|
| 5 |
+
"train_samples_per_second": 13.96,
|
| 6 |
+
"train_steps_per_second": 0.872
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"train_loss": 1.6477281238721764,
|
| 4 |
+
"train_runtime": 4221.2735,
|
| 5 |
+
"train_samples_per_second": 13.96,
|
| 6 |
+
"train_steps_per_second": 0.872
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,136 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.9967417865870214,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3680,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.54,
|
| 13 |
+
"learning_rate": 1.731521739130435e-05,
|
| 14 |
+
"loss": 1.9542,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 1.0,
|
| 19 |
+
"eval_f1": 0.9149,
|
| 20 |
+
"eval_gen_len": 18.56723716381418,
|
| 21 |
+
"eval_loss": 1.534995198249817,
|
| 22 |
+
"eval_precision": 0.9229,
|
| 23 |
+
"eval_recall": 0.9074,
|
| 24 |
+
"eval_rouge1": 0.4928,
|
| 25 |
+
"eval_rouge2": 0.2436,
|
| 26 |
+
"eval_rougeL": 0.4085,
|
| 27 |
+
"eval_rougeLsum": 0.4086,
|
| 28 |
+
"eval_runtime": 246.8012,
|
| 29 |
+
"eval_samples_per_second": 3.314,
|
| 30 |
+
"eval_steps_per_second": 0.831,
|
| 31 |
+
"step": 920
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.09,
|
| 35 |
+
"learning_rate": 1.4603260869565218e-05,
|
| 36 |
+
"loss": 1.7174,
|
| 37 |
+
"step": 1000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.63,
|
| 41 |
+
"learning_rate": 1.1885869565217392e-05,
|
| 42 |
+
"loss": 1.6331,
|
| 43 |
+
"step": 1500
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"eval_f1": 0.9166,
|
| 48 |
+
"eval_gen_len": 18.815403422982886,
|
| 49 |
+
"eval_loss": 1.491409182548523,
|
| 50 |
+
"eval_precision": 0.9246,
|
| 51 |
+
"eval_recall": 0.9092,
|
| 52 |
+
"eval_rouge1": 0.5037,
|
| 53 |
+
"eval_rouge2": 0.257,
|
| 54 |
+
"eval_rougeL": 0.4202,
|
| 55 |
+
"eval_rougeLsum": 0.4206,
|
| 56 |
+
"eval_runtime": 247.8681,
|
| 57 |
+
"eval_samples_per_second": 3.3,
|
| 58 |
+
"eval_steps_per_second": 0.827,
|
| 59 |
+
"step": 1841
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.17,
|
| 63 |
+
"learning_rate": 9.179347826086958e-06,
|
| 64 |
+
"loss": 1.5996,
|
| 65 |
+
"step": 2000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.72,
|
| 69 |
+
"learning_rate": 6.472826086956522e-06,
|
| 70 |
+
"loss": 1.5694,
|
| 71 |
+
"step": 2500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.0,
|
| 75 |
+
"eval_f1": 0.917,
|
| 76 |
+
"eval_gen_len": 19.448655256723715,
|
| 77 |
+
"eval_loss": 1.4760992527008057,
|
| 78 |
+
"eval_precision": 0.9241,
|
| 79 |
+
"eval_recall": 0.9103,
|
| 80 |
+
"eval_rouge1": 0.5071,
|
| 81 |
+
"eval_rouge2": 0.259,
|
| 82 |
+
"eval_rougeL": 0.4212,
|
| 83 |
+
"eval_rougeLsum": 0.4214,
|
| 84 |
+
"eval_runtime": 249.775,
|
| 85 |
+
"eval_samples_per_second": 3.275,
|
| 86 |
+
"eval_steps_per_second": 0.821,
|
| 87 |
+
"step": 2762
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 3.26,
|
| 91 |
+
"learning_rate": 3.7554347826086963e-06,
|
| 92 |
+
"loss": 1.5609,
|
| 93 |
+
"step": 3000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 3.8,
|
| 97 |
+
"learning_rate": 1.048913043478261e-06,
|
| 98 |
+
"loss": 1.5374,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.0,
|
| 103 |
+
"eval_f1": 0.917,
|
| 104 |
+
"eval_gen_len": 19.1479217603912,
|
| 105 |
+
"eval_loss": 1.4708906412124634,
|
| 106 |
+
"eval_precision": 0.9247,
|
| 107 |
+
"eval_recall": 0.9099,
|
| 108 |
+
"eval_rouge1": 0.5072,
|
| 109 |
+
"eval_rouge2": 0.2631,
|
| 110 |
+
"eval_rougeL": 0.4243,
|
| 111 |
+
"eval_rougeLsum": 0.4244,
|
| 112 |
+
"eval_runtime": 248.4801,
|
| 113 |
+
"eval_samples_per_second": 3.292,
|
| 114 |
+
"eval_steps_per_second": 0.825,
|
| 115 |
+
"step": 3680
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 4.0,
|
| 119 |
+
"step": 3680,
|
| 120 |
+
"total_flos": 3.997817815616717e+16,
|
| 121 |
+
"train_loss": 1.6477281238721764,
|
| 122 |
+
"train_runtime": 4221.2735,
|
| 123 |
+
"train_samples_per_second": 13.96,
|
| 124 |
+
"train_steps_per_second": 0.872
|
| 125 |
+
}
|
| 126 |
+
],
|
| 127 |
+
"logging_steps": 500,
|
| 128 |
+
"max_steps": 3680,
|
| 129 |
+
"num_input_tokens_seen": 0,
|
| 130 |
+
"num_train_epochs": 4,
|
| 131 |
+
"save_steps": 500,
|
| 132 |
+
"total_flos": 3.997817815616717e+16,
|
| 133 |
+
"train_batch_size": 4,
|
| 134 |
+
"trial_name": null,
|
| 135 |
+
"trial_params": null
|
| 136 |
+
}
|