Training in progress, step 400
Browse files- all_results.json +20 -0
- eval_results.json +14 -0
- model.safetensors +1 -1
- train_results.json +9 -0
- trainer_state.json +472 -0
- training_args.bin +1 -1
all_results.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.259558164180321,
|
| 3 |
+
"eval_accuracy": 0.5679713544882085,
|
| 4 |
+
"eval_f1_macro": 0.535328565889324,
|
| 5 |
+
"eval_f1_micro": 0.5679713544882085,
|
| 6 |
+
"eval_loss": 0.45331016182899475,
|
| 7 |
+
"eval_precision_macro": 0.5489792373431622,
|
| 8 |
+
"eval_recall_macro": 0.5315336191512473,
|
| 9 |
+
"eval_roc_auc": 0.786883264678242,
|
| 10 |
+
"eval_runtime": 53.0285,
|
| 11 |
+
"eval_samples": 8099,
|
| 12 |
+
"eval_samples_per_second": 152.729,
|
| 13 |
+
"eval_steps_per_second": 76.374,
|
| 14 |
+
"total_flos": 1.052068446713856e+16,
|
| 15 |
+
"train_loss": 0.48495769653320314,
|
| 16 |
+
"train_runtime": 3435.9427,
|
| 17 |
+
"train_samples": 24534,
|
| 18 |
+
"train_samples_per_second": 285.616,
|
| 19 |
+
"train_steps_per_second": 17.858
|
| 20 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.259558164180321,
|
| 3 |
+
"eval_accuracy": 0.5679713544882085,
|
| 4 |
+
"eval_f1_macro": 0.535328565889324,
|
| 5 |
+
"eval_f1_micro": 0.5679713544882085,
|
| 6 |
+
"eval_loss": 0.45331016182899475,
|
| 7 |
+
"eval_precision_macro": 0.5489792373431622,
|
| 8 |
+
"eval_recall_macro": 0.5315336191512473,
|
| 9 |
+
"eval_roc_auc": 0.786883264678242,
|
| 10 |
+
"eval_runtime": 53.0285,
|
| 11 |
+
"eval_samples": 8099,
|
| 12 |
+
"eval_samples_per_second": 152.729,
|
| 13 |
+
"eval_steps_per_second": 76.374
|
| 14 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 498619448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f390e449ff5db0a2a317b8dc0d2aad5c00cc7335717d5811bda00ae8aec69b8c
|
| 3 |
size 498619448
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.259558164180321,
|
| 3 |
+
"total_flos": 1.052068446713856e+16,
|
| 4 |
+
"train_loss": 0.48495769653320314,
|
| 5 |
+
"train_runtime": 3435.9427,
|
| 6 |
+
"train_samples": 24534,
|
| 7 |
+
"train_samples_per_second": 285.616,
|
| 8 |
+
"train_steps_per_second": 17.858
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,472 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 4400,
|
| 3 |
+
"best_metric": 0.453235387802124,
|
| 4 |
+
"best_model_checkpoint": "roberta-base-multi-head-4-directions/checkpoint-4400",
|
| 5 |
+
"epoch": 3.259558164180321,
|
| 6 |
+
"eval_steps": 200,
|
| 7 |
+
"global_step": 5000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.13043123828156844,
|
| 14 |
+
"eval_accuracy": 0.15051240893937523,
|
| 15 |
+
"eval_f1_macro": 0.0654110324103885,
|
| 16 |
+
"eval_f1_micro": 0.15051240893937523,
|
| 17 |
+
"eval_loss": 0.6699725985527039,
|
| 18 |
+
"eval_precision_macro": 0.03762810223484381,
|
| 19 |
+
"eval_recall_macro": 0.25,
|
| 20 |
+
"eval_roc_auc": 0.5065070998981289,
|
| 21 |
+
"eval_runtime": 58.3231,
|
| 22 |
+
"eval_samples_per_second": 138.864,
|
| 23 |
+
"eval_steps_per_second": 69.441,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.2608624765631369,
|
| 28 |
+
"eval_accuracy": 0.3306581059390048,
|
| 29 |
+
"eval_f1_macro": 0.21851269600524115,
|
| 30 |
+
"eval_f1_micro": 0.3306581059390048,
|
| 31 |
+
"eval_loss": 0.5907890796661377,
|
| 32 |
+
"eval_precision_macro": 0.3477712230149137,
|
| 33 |
+
"eval_recall_macro": 0.2628129027541947,
|
| 34 |
+
"eval_roc_auc": 0.5264776449526523,
|
| 35 |
+
"eval_runtime": 58.515,
|
| 36 |
+
"eval_samples_per_second": 138.409,
|
| 37 |
+
"eval_steps_per_second": 69.213,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.3260780957039211,
|
| 42 |
+
"grad_norm": 1.910775899887085,
|
| 43 |
+
"learning_rate": 8.132333767926989e-07,
|
| 44 |
+
"loss": 0.6418,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.3912937148447053,
|
| 49 |
+
"eval_accuracy": 0.39646869983948635,
|
| 50 |
+
"eval_f1_macro": 0.24202854764905962,
|
| 51 |
+
"eval_f1_micro": 0.39646869983948635,
|
| 52 |
+
"eval_loss": 0.5681799650192261,
|
| 53 |
+
"eval_precision_macro": 0.31314716750672944,
|
| 54 |
+
"eval_recall_macro": 0.29579537486279056,
|
| 55 |
+
"eval_roc_auc": 0.5825887308747437,
|
| 56 |
+
"eval_runtime": 58.253,
|
| 57 |
+
"eval_samples_per_second": 139.031,
|
| 58 |
+
"eval_steps_per_second": 69.524,
|
| 59 |
+
"step": 600
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.5217249531262738,
|
| 63 |
+
"eval_accuracy": 0.4446227929373997,
|
| 64 |
+
"eval_f1_macro": 0.2752091664099515,
|
| 65 |
+
"eval_f1_micro": 0.4446227929373997,
|
| 66 |
+
"eval_loss": 0.5510929226875305,
|
| 67 |
+
"eval_precision_macro": 0.3725228720549576,
|
| 68 |
+
"eval_recall_macro": 0.3347772936726068,
|
| 69 |
+
"eval_roc_auc": 0.6511260809997589,
|
| 70 |
+
"eval_runtime": 58.5957,
|
| 71 |
+
"eval_samples_per_second": 138.218,
|
| 72 |
+
"eval_steps_per_second": 69.118,
|
| 73 |
+
"step": 800
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.6521561914078422,
|
| 77 |
+
"grad_norm": 3.2853586673736572,
|
| 78 |
+
"learning_rate": 1.6280964797913951e-06,
|
| 79 |
+
"loss": 0.56,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.6521561914078422,
|
| 84 |
+
"eval_accuracy": 0.47956537844178293,
|
| 85 |
+
"eval_f1_macro": 0.339724330856578,
|
| 86 |
+
"eval_f1_micro": 0.47956537844178293,
|
| 87 |
+
"eval_loss": 0.5183268189430237,
|
| 88 |
+
"eval_precision_macro": 0.5998469679206377,
|
| 89 |
+
"eval_recall_macro": 0.37515425621657034,
|
| 90 |
+
"eval_roc_auc": 0.7004256590184301,
|
| 91 |
+
"eval_runtime": 58.4596,
|
| 92 |
+
"eval_samples_per_second": 138.54,
|
| 93 |
+
"eval_steps_per_second": 69.279,
|
| 94 |
+
"step": 1000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.7825874296894106,
|
| 98 |
+
"eval_accuracy": 0.4951228546734165,
|
| 99 |
+
"eval_f1_macro": 0.3960300119997172,
|
| 100 |
+
"eval_f1_micro": 0.4951228546734165,
|
| 101 |
+
"eval_loss": 0.5027472972869873,
|
| 102 |
+
"eval_precision_macro": 0.47649685990985047,
|
| 103 |
+
"eval_recall_macro": 0.42432070581801545,
|
| 104 |
+
"eval_roc_auc": 0.7210076403173139,
|
| 105 |
+
"eval_runtime": 59.2756,
|
| 106 |
+
"eval_samples_per_second": 136.633,
|
| 107 |
+
"eval_steps_per_second": 68.325,
|
| 108 |
+
"step": 1200
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.9130186679709791,
|
| 112 |
+
"eval_accuracy": 0.49364119027040376,
|
| 113 |
+
"eval_f1_macro": 0.4202868285905695,
|
| 114 |
+
"eval_f1_micro": 0.49364119027040376,
|
| 115 |
+
"eval_loss": 0.5012013912200928,
|
| 116 |
+
"eval_precision_macro": 0.476097069319352,
|
| 117 |
+
"eval_recall_macro": 0.44606990421659504,
|
| 118 |
+
"eval_roc_auc": 0.7341331293715618,
|
| 119 |
+
"eval_runtime": 58.8986,
|
| 120 |
+
"eval_samples_per_second": 137.507,
|
| 121 |
+
"eval_steps_per_second": 68.762,
|
| 122 |
+
"step": 1400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.9782342871117633,
|
| 126 |
+
"grad_norm": 6.466639041900635,
|
| 127 |
+
"learning_rate": 2.4429595827900914e-06,
|
| 128 |
+
"loss": 0.5136,
|
| 129 |
+
"step": 1500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.0430423086329177,
|
| 133 |
+
"eval_accuracy": 0.5227805901963205,
|
| 134 |
+
"eval_f1_macro": 0.4272500725974152,
|
| 135 |
+
"eval_f1_micro": 0.5227805901963205,
|
| 136 |
+
"eval_loss": 0.4897969961166382,
|
| 137 |
+
"eval_precision_macro": 0.511088683081628,
|
| 138 |
+
"eval_recall_macro": 0.4450094829273592,
|
| 139 |
+
"eval_roc_auc": 0.7377316174079231,
|
| 140 |
+
"eval_runtime": 58.3834,
|
| 141 |
+
"eval_samples_per_second": 138.721,
|
| 142 |
+
"eval_steps_per_second": 69.369,
|
| 143 |
+
"step": 1600
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.1734735469144861,
|
| 147 |
+
"eval_accuracy": 0.5304358562785529,
|
| 148 |
+
"eval_f1_macro": 0.47228705140051763,
|
| 149 |
+
"eval_f1_micro": 0.5304358562785529,
|
| 150 |
+
"eval_loss": 0.48175758123397827,
|
| 151 |
+
"eval_precision_macro": 0.49865343258508565,
|
| 152 |
+
"eval_recall_macro": 0.4751571425140743,
|
| 153 |
+
"eval_roc_auc": 0.7504044589822995,
|
| 154 |
+
"eval_runtime": 59.0784,
|
| 155 |
+
"eval_samples_per_second": 137.089,
|
| 156 |
+
"eval_steps_per_second": 68.553,
|
| 157 |
+
"step": 1800
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.3039047851960546,
|
| 161 |
+
"grad_norm": 7.449667453765869,
|
| 162 |
+
"learning_rate": 3.257822685788788e-06,
|
| 163 |
+
"loss": 0.4842,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.3039047851960546,
|
| 168 |
+
"eval_accuracy": 0.5330287689838251,
|
| 169 |
+
"eval_f1_macro": 0.49309253039408907,
|
| 170 |
+
"eval_f1_micro": 0.5330287689838251,
|
| 171 |
+
"eval_loss": 0.4809001386165619,
|
| 172 |
+
"eval_precision_macro": 0.5034709810584846,
|
| 173 |
+
"eval_recall_macro": 0.4966387871579345,
|
| 174 |
+
"eval_roc_auc": 0.758014108468557,
|
| 175 |
+
"eval_runtime": 58.3487,
|
| 176 |
+
"eval_samples_per_second": 138.803,
|
| 177 |
+
"eval_steps_per_second": 69.41,
|
| 178 |
+
"step": 2000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.434336023477623,
|
| 182 |
+
"eval_accuracy": 0.5411779232003952,
|
| 183 |
+
"eval_f1_macro": 0.5090942441371813,
|
| 184 |
+
"eval_f1_micro": 0.5411779232003952,
|
| 185 |
+
"eval_loss": 0.47537535429000854,
|
| 186 |
+
"eval_precision_macro": 0.5108825390909519,
|
| 187 |
+
"eval_recall_macro": 0.5109581830122815,
|
| 188 |
+
"eval_roc_auc": 0.7651162037942221,
|
| 189 |
+
"eval_runtime": 58.7575,
|
| 190 |
+
"eval_samples_per_second": 137.838,
|
| 191 |
+
"eval_steps_per_second": 68.927,
|
| 192 |
+
"step": 2200
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.5647672617591915,
|
| 196 |
+
"eval_accuracy": 0.5522904062229905,
|
| 197 |
+
"eval_f1_macro": 0.5071849375719386,
|
| 198 |
+
"eval_f1_micro": 0.5522904062229905,
|
| 199 |
+
"eval_loss": 0.468904048204422,
|
| 200 |
+
"eval_precision_macro": 0.5307207892146724,
|
| 201 |
+
"eval_recall_macro": 0.5061303129931647,
|
| 202 |
+
"eval_roc_auc": 0.7695441321425415,
|
| 203 |
+
"eval_runtime": 58.5621,
|
| 204 |
+
"eval_samples_per_second": 138.298,
|
| 205 |
+
"eval_steps_per_second": 69.157,
|
| 206 |
+
"step": 2400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.6299828808999757,
|
| 210 |
+
"grad_norm": 6.51912260055542,
|
| 211 |
+
"learning_rate": 4.072685788787484e-06,
|
| 212 |
+
"loss": 0.4695,
|
| 213 |
+
"step": 2500
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.69519850004076,
|
| 217 |
+
"eval_accuracy": 0.5379676503272008,
|
| 218 |
+
"eval_f1_macro": 0.48087910671865897,
|
| 219 |
+
"eval_f1_micro": 0.5379676503272008,
|
| 220 |
+
"eval_loss": 0.4785490036010742,
|
| 221 |
+
"eval_precision_macro": 0.5346246897655283,
|
| 222 |
+
"eval_recall_macro": 0.4906654493419952,
|
| 223 |
+
"eval_roc_auc": 0.7663202674809741,
|
| 224 |
+
"eval_runtime": 58.4801,
|
| 225 |
+
"eval_samples_per_second": 138.491,
|
| 226 |
+
"eval_steps_per_second": 69.254,
|
| 227 |
+
"step": 2600
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.8256297383223283,
|
| 231 |
+
"eval_accuracy": 0.5545129028275095,
|
| 232 |
+
"eval_f1_macro": 0.5144296310304717,
|
| 233 |
+
"eval_f1_micro": 0.5545129028275095,
|
| 234 |
+
"eval_loss": 0.46492108702659607,
|
| 235 |
+
"eval_precision_macro": 0.5301331529844335,
|
| 236 |
+
"eval_recall_macro": 0.5080649601410273,
|
| 237 |
+
"eval_roc_auc": 0.7745130299239165,
|
| 238 |
+
"eval_runtime": 58.563,
|
| 239 |
+
"eval_samples_per_second": 138.295,
|
| 240 |
+
"eval_steps_per_second": 69.156,
|
| 241 |
+
"step": 2800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.9560609766038968,
|
| 245 |
+
"grad_norm": 6.679250717163086,
|
| 246 |
+
"learning_rate": 4.88754889178618e-06,
|
| 247 |
+
"loss": 0.4655,
|
| 248 |
+
"step": 3000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.9560609766038968,
|
| 252 |
+
"eval_accuracy": 0.5494505494505495,
|
| 253 |
+
"eval_f1_macro": 0.5210019068338745,
|
| 254 |
+
"eval_f1_micro": 0.5494505494505495,
|
| 255 |
+
"eval_loss": 0.46518564224243164,
|
| 256 |
+
"eval_precision_macro": 0.5276079708493782,
|
| 257 |
+
"eval_recall_macro": 0.5245104582824609,
|
| 258 |
+
"eval_roc_auc": 0.7757459310311635,
|
| 259 |
+
"eval_runtime": 58.703,
|
| 260 |
+
"eval_samples_per_second": 137.966,
|
| 261 |
+
"eval_steps_per_second": 68.991,
|
| 262 |
+
"step": 3000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 2.0860846172658354,
|
| 266 |
+
"eval_accuracy": 0.5653784417829362,
|
| 267 |
+
"eval_f1_macro": 0.528637637687615,
|
| 268 |
+
"eval_f1_micro": 0.5653784417829362,
|
| 269 |
+
"eval_loss": 0.46096640825271606,
|
| 270 |
+
"eval_precision_macro": 0.5443547583807372,
|
| 271 |
+
"eval_recall_macro": 0.5223821110706193,
|
| 272 |
+
"eval_roc_auc": 0.781316894157911,
|
| 273 |
+
"eval_runtime": 58.6208,
|
| 274 |
+
"eval_samples_per_second": 138.159,
|
| 275 |
+
"eval_steps_per_second": 69.088,
|
| 276 |
+
"step": 3200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 2.216515855547404,
|
| 280 |
+
"eval_accuracy": 0.5498209655513027,
|
| 281 |
+
"eval_f1_macro": 0.5186060540296338,
|
| 282 |
+
"eval_f1_micro": 0.5498209655513027,
|
| 283 |
+
"eval_loss": 0.46535903215408325,
|
| 284 |
+
"eval_precision_macro": 0.5399973417347204,
|
| 285 |
+
"eval_recall_macro": 0.5255215009903015,
|
| 286 |
+
"eval_roc_auc": 0.7808947543899514,
|
| 287 |
+
"eval_runtime": 58.434,
|
| 288 |
+
"eval_samples_per_second": 138.601,
|
| 289 |
+
"eval_steps_per_second": 69.309,
|
| 290 |
+
"step": 3400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.281731474688188,
|
| 294 |
+
"grad_norm": 7.935737609863281,
|
| 295 |
+
"learning_rate": 4.963030947642902e-06,
|
| 296 |
+
"loss": 0.4414,
|
| 297 |
+
"step": 3500
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 2.3469470938289723,
|
| 301 |
+
"eval_accuracy": 0.5530312384244969,
|
| 302 |
+
"eval_f1_macro": 0.5192037737144601,
|
| 303 |
+
"eval_f1_micro": 0.5530312384244969,
|
| 304 |
+
"eval_loss": 0.4659627676010132,
|
| 305 |
+
"eval_precision_macro": 0.542720199525297,
|
| 306 |
+
"eval_recall_macro": 0.5266595526270865,
|
| 307 |
+
"eval_roc_auc": 0.7817212427653404,
|
| 308 |
+
"eval_runtime": 58.8426,
|
| 309 |
+
"eval_samples_per_second": 137.638,
|
| 310 |
+
"eval_steps_per_second": 68.828,
|
| 311 |
+
"step": 3600
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 2.4773783321105407,
|
| 315 |
+
"eval_accuracy": 0.564390665514261,
|
| 316 |
+
"eval_f1_macro": 0.5393273157888367,
|
| 317 |
+
"eval_f1_micro": 0.564390665514261,
|
| 318 |
+
"eval_loss": 0.45927414298057556,
|
| 319 |
+
"eval_precision_macro": 0.5430622655929049,
|
| 320 |
+
"eval_recall_macro": 0.5388167782580582,
|
| 321 |
+
"eval_roc_auc": 0.785497056666471,
|
| 322 |
+
"eval_runtime": 60.3532,
|
| 323 |
+
"eval_samples_per_second": 134.193,
|
| 324 |
+
"eval_steps_per_second": 67.105,
|
| 325 |
+
"step": 3800
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 2.607809570392109,
|
| 329 |
+
"grad_norm": 6.252172946929932,
|
| 330 |
+
"learning_rate": 4.920143415906128e-06,
|
| 331 |
+
"loss": 0.433,
|
| 332 |
+
"step": 4000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.607809570392109,
|
| 336 |
+
"eval_accuracy": 0.5583405358686258,
|
| 337 |
+
"eval_f1_macro": 0.5308754875566113,
|
| 338 |
+
"eval_f1_micro": 0.5583405358686258,
|
| 339 |
+
"eval_loss": 0.4604816436767578,
|
| 340 |
+
"eval_precision_macro": 0.5405007742715062,
|
| 341 |
+
"eval_recall_macro": 0.5347270499333006,
|
| 342 |
+
"eval_roc_auc": 0.7849083537765603,
|
| 343 |
+
"eval_runtime": 58.4178,
|
| 344 |
+
"eval_samples_per_second": 138.639,
|
| 345 |
+
"eval_steps_per_second": 69.328,
|
| 346 |
+
"step": 4000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 2.7382408086736776,
|
| 350 |
+
"eval_accuracy": 0.5485862452154587,
|
| 351 |
+
"eval_f1_macro": 0.5244598378559915,
|
| 352 |
+
"eval_f1_micro": 0.5485862452154587,
|
| 353 |
+
"eval_loss": 0.46733224391937256,
|
| 354 |
+
"eval_precision_macro": 0.5429571340400278,
|
| 355 |
+
"eval_recall_macro": 0.5359068311148909,
|
| 356 |
+
"eval_roc_auc": 0.7850860010461288,
|
| 357 |
+
"eval_runtime": 58.5291,
|
| 358 |
+
"eval_samples_per_second": 138.376,
|
| 359 |
+
"eval_steps_per_second": 69.196,
|
| 360 |
+
"step": 4200
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 2.868672046955246,
|
| 364 |
+
"eval_accuracy": 0.5685887146561304,
|
| 365 |
+
"eval_f1_macro": 0.5358137721542776,
|
| 366 |
+
"eval_f1_micro": 0.5685887146561304,
|
| 367 |
+
"eval_loss": 0.453235387802124,
|
| 368 |
+
"eval_precision_macro": 0.5498552760907015,
|
| 369 |
+
"eval_recall_macro": 0.531919882945018,
|
| 370 |
+
"eval_roc_auc": 0.7869039137806253,
|
| 371 |
+
"eval_runtime": 58.2827,
|
| 372 |
+
"eval_samples_per_second": 138.961,
|
| 373 |
+
"eval_steps_per_second": 69.489,
|
| 374 |
+
"step": 4400
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.9338876660960302,
|
| 378 |
+
"grad_norm": 7.882796287536621,
|
| 379 |
+
"learning_rate": 4.8772558841693544e-06,
|
| 380 |
+
"loss": 0.435,
|
| 381 |
+
"step": 4500
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.9991032852368145,
|
| 385 |
+
"eval_accuracy": 0.5695764909248056,
|
| 386 |
+
"eval_f1_macro": 0.5394199448286808,
|
| 387 |
+
"eval_f1_micro": 0.5695764909248056,
|
| 388 |
+
"eval_loss": 0.4586756229400635,
|
| 389 |
+
"eval_precision_macro": 0.5469928436264997,
|
| 390 |
+
"eval_recall_macro": 0.5385537439613218,
|
| 391 |
+
"eval_roc_auc": 0.7865007504045092,
|
| 392 |
+
"eval_runtime": 58.5001,
|
| 393 |
+
"eval_samples_per_second": 138.444,
|
| 394 |
+
"eval_steps_per_second": 69.231,
|
| 395 |
+
"step": 4600
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 3.1291269258987526,
|
| 399 |
+
"eval_accuracy": 0.5659958019508581,
|
| 400 |
+
"eval_f1_macro": 0.5352002738500827,
|
| 401 |
+
"eval_f1_micro": 0.5659958019508581,
|
| 402 |
+
"eval_loss": 0.46012088656425476,
|
| 403 |
+
"eval_precision_macro": 0.5485092746099708,
|
| 404 |
+
"eval_recall_macro": 0.537293222573151,
|
| 405 |
+
"eval_roc_auc": 0.7868513938139832,
|
| 406 |
+
"eval_runtime": 58.3581,
|
| 407 |
+
"eval_samples_per_second": 138.781,
|
| 408 |
+
"eval_steps_per_second": 69.399,
|
| 409 |
+
"step": 4800
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 3.259558164180321,
|
| 413 |
+
"grad_norm": 6.890503406524658,
|
| 414 |
+
"learning_rate": 4.834368352432582e-06,
|
| 415 |
+
"loss": 0.4056,
|
| 416 |
+
"step": 5000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 3.259558164180321,
|
| 420 |
+
"eval_accuracy": 0.5674774663538709,
|
| 421 |
+
"eval_f1_macro": 0.5441156714097114,
|
| 422 |
+
"eval_f1_micro": 0.5674774663538709,
|
| 423 |
+
"eval_loss": 0.47018805146217346,
|
| 424 |
+
"eval_precision_macro": 0.5454139749449086,
|
| 425 |
+
"eval_recall_macro": 0.5461421293276678,
|
| 426 |
+
"eval_roc_auc": 0.7877411895075941,
|
| 427 |
+
"eval_runtime": 58.8315,
|
| 428 |
+
"eval_samples_per_second": 137.664,
|
| 429 |
+
"eval_steps_per_second": 68.841,
|
| 430 |
+
"step": 5000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 3.259558164180321,
|
| 434 |
+
"step": 5000,
|
| 435 |
+
"total_flos": 1.052068446713856e+16,
|
| 436 |
+
"train_loss": 0.48495769653320314,
|
| 437 |
+
"train_runtime": 3435.9427,
|
| 438 |
+
"train_samples_per_second": 285.616,
|
| 439 |
+
"train_steps_per_second": 17.858
|
| 440 |
+
}
|
| 441 |
+
],
|
| 442 |
+
"logging_steps": 500,
|
| 443 |
+
"max_steps": 61360,
|
| 444 |
+
"num_input_tokens_seen": 0,
|
| 445 |
+
"num_train_epochs": 40,
|
| 446 |
+
"save_steps": 400,
|
| 447 |
+
"stateful_callbacks": {
|
| 448 |
+
"EarlyStoppingCallback": {
|
| 449 |
+
"args": {
|
| 450 |
+
"early_stopping_patience": 3,
|
| 451 |
+
"early_stopping_threshold": 0.0
|
| 452 |
+
},
|
| 453 |
+
"attributes": {
|
| 454 |
+
"early_stopping_patience_counter": 2
|
| 455 |
+
}
|
| 456 |
+
},
|
| 457 |
+
"TrainerControl": {
|
| 458 |
+
"args": {
|
| 459 |
+
"should_epoch_stop": false,
|
| 460 |
+
"should_evaluate": false,
|
| 461 |
+
"should_log": false,
|
| 462 |
+
"should_save": true,
|
| 463 |
+
"should_training_stop": false
|
| 464 |
+
},
|
| 465 |
+
"attributes": {}
|
| 466 |
+
}
|
| 467 |
+
},
|
| 468 |
+
"total_flos": 1.052068446713856e+16,
|
| 469 |
+
"train_batch_size": 2,
|
| 470 |
+
"trial_name": null,
|
| 471 |
+
"trial_params": null
|
| 472 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5432
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98340f0082a087f7f1a653b3d6ce4880ec594bfdeda51ab5b5528703929f2520
|
| 3 |
size 5432
|