End of training
Browse files- README.md +5 -5
- all_results.json +16 -0
- test_results.json +11 -0
- train_results.json +8 -0
- trainer_state.json +429 -0
README.md
CHANGED
|
@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [vinai/phobert-base-v2](https://huggingface.co/vinai/phobert-base-v2) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
-
- F1 Score: 0.
|
| 24 |
-
- Recall: 0.
|
| 25 |
-
- Precision: 0.
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [vinai/phobert-base-v2](https://huggingface.co/vinai/phobert-base-v2) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.6228
|
| 22 |
+
- Accuracy: 0.9272
|
| 23 |
+
- F1 Score: 0.8944
|
| 24 |
+
- Recall: 0.8843
|
| 25 |
+
- Precision: 0.9063
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 21.88,
|
| 3 |
+
"eval_accuracy": 0.9271955599085864,
|
| 4 |
+
"eval_f1_score": 0.8943646614708781,
|
| 5 |
+
"eval_loss": 0.6228047013282776,
|
| 6 |
+
"eval_precision": 0.9063376648453086,
|
| 7 |
+
"eval_recall": 0.8842502417202245,
|
| 8 |
+
"eval_runtime": 4.7903,
|
| 9 |
+
"eval_samples_per_second": 639.416,
|
| 10 |
+
"eval_steps_per_second": 10.02,
|
| 11 |
+
"total_flos": 4711306254898320.0,
|
| 12 |
+
"train_loss": 0.5936482783726283,
|
| 13 |
+
"train_runtime": 1811.9756,
|
| 14 |
+
"train_samples_per_second": 282.565,
|
| 15 |
+
"train_steps_per_second": 2.208
|
| 16 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 21.88,
|
| 3 |
+
"eval_accuracy": 0.9271955599085864,
|
| 4 |
+
"eval_f1_score": 0.8943646614708781,
|
| 5 |
+
"eval_loss": 0.6228047013282776,
|
| 6 |
+
"eval_precision": 0.9063376648453086,
|
| 7 |
+
"eval_recall": 0.8842502417202245,
|
| 8 |
+
"eval_runtime": 4.7903,
|
| 9 |
+
"eval_samples_per_second": 639.416,
|
| 10 |
+
"eval_steps_per_second": 10.02
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 21.88,
|
| 3 |
+
"total_flos": 4711306254898320.0,
|
| 4 |
+
"train_loss": 0.5936482783726283,
|
| 5 |
+
"train_runtime": 1811.9756,
|
| 6 |
+
"train_samples_per_second": 282.565,
|
| 7 |
+
"train_steps_per_second": 2.208
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,429 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8943646614708781,
|
| 3 |
+
"best_model_checkpoint": "cls_comment-phobert-base-v2-v2.3/checkpoint-1600",
|
| 4 |
+
"epoch": 21.875,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 2100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.04,
|
| 13 |
+
"grad_norm": 2.3460028171539307,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 1.1884,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.04,
|
| 20 |
+
"eval_accuracy": 0.7277179236043095,
|
| 21 |
+
"eval_f1_score": 0.43708201591573076,
|
| 22 |
+
"eval_loss": 1.0905314683914185,
|
| 23 |
+
"eval_precision": 0.5016943776884798,
|
| 24 |
+
"eval_recall": 0.42288884208445504,
|
| 25 |
+
"eval_runtime": 4.9336,
|
| 26 |
+
"eval_samples_per_second": 620.848,
|
| 27 |
+
"eval_steps_per_second": 9.729,
|
| 28 |
+
"step": 100
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.08,
|
| 32 |
+
"grad_norm": 9.137602806091309,
|
| 33 |
+
"learning_rate": 5e-06,
|
| 34 |
+
"loss": 1.0502,
|
| 35 |
+
"step": 200
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.08,
|
| 39 |
+
"eval_accuracy": 0.7956252040483186,
|
| 40 |
+
"eval_f1_score": 0.5245336258309143,
|
| 41 |
+
"eval_loss": 0.9278329014778137,
|
| 42 |
+
"eval_precision": 0.5244123326675032,
|
| 43 |
+
"eval_recall": 0.5273852871687749,
|
| 44 |
+
"eval_runtime": 4.9622,
|
| 45 |
+
"eval_samples_per_second": 617.266,
|
| 46 |
+
"eval_steps_per_second": 9.673,
|
| 47 |
+
"step": 200
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.12,
|
| 51 |
+
"grad_norm": 8.277727127075195,
|
| 52 |
+
"learning_rate": 7.500000000000001e-06,
|
| 53 |
+
"loss": 0.9031,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.12,
|
| 58 |
+
"eval_accuracy": 0.8445968005223637,
|
| 59 |
+
"eval_f1_score": 0.5653211085059351,
|
| 60 |
+
"eval_loss": 0.7793412208557129,
|
| 61 |
+
"eval_precision": 0.7067057672713553,
|
| 62 |
+
"eval_recall": 0.589697230912186,
|
| 63 |
+
"eval_runtime": 5.0159,
|
| 64 |
+
"eval_samples_per_second": 610.664,
|
| 65 |
+
"eval_steps_per_second": 9.57,
|
| 66 |
+
"step": 300
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.17,
|
| 70 |
+
"grad_norm": 3.594104290008545,
|
| 71 |
+
"learning_rate": 1e-05,
|
| 72 |
+
"loss": 0.7772,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.17,
|
| 77 |
+
"eval_accuracy": 0.8697355533790402,
|
| 78 |
+
"eval_f1_score": 0.6917446533625714,
|
| 79 |
+
"eval_loss": 0.7089904546737671,
|
| 80 |
+
"eval_precision": 0.7268852070004593,
|
| 81 |
+
"eval_recall": 0.6804440791993415,
|
| 82 |
+
"eval_runtime": 4.9399,
|
| 83 |
+
"eval_samples_per_second": 620.058,
|
| 84 |
+
"eval_steps_per_second": 9.717,
|
| 85 |
+
"step": 400
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 5.21,
|
| 89 |
+
"grad_norm": 4.503917694091797,
|
| 90 |
+
"learning_rate": 9.722222222222223e-06,
|
| 91 |
+
"loss": 0.6741,
|
| 92 |
+
"step": 500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.21,
|
| 96 |
+
"eval_accuracy": 0.8958537381651975,
|
| 97 |
+
"eval_f1_score": 0.8078466525990172,
|
| 98 |
+
"eval_loss": 0.648346483707428,
|
| 99 |
+
"eval_precision": 0.8797678600731677,
|
| 100 |
+
"eval_recall": 0.7817903123125977,
|
| 101 |
+
"eval_runtime": 4.9504,
|
| 102 |
+
"eval_samples_per_second": 618.741,
|
| 103 |
+
"eval_steps_per_second": 9.696,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.25,
|
| 108 |
+
"grad_norm": 4.407541275024414,
|
| 109 |
+
"learning_rate": 9.444444444444445e-06,
|
| 110 |
+
"loss": 0.6156,
|
| 111 |
+
"step": 600
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 6.25,
|
| 115 |
+
"eval_accuracy": 0.9079333986287953,
|
| 116 |
+
"eval_f1_score": 0.8601153974079928,
|
| 117 |
+
"eval_loss": 0.6268225908279419,
|
| 118 |
+
"eval_precision": 0.8933059224979765,
|
| 119 |
+
"eval_recall": 0.8344643413921417,
|
| 120 |
+
"eval_runtime": 4.9864,
|
| 121 |
+
"eval_samples_per_second": 614.266,
|
| 122 |
+
"eval_steps_per_second": 9.626,
|
| 123 |
+
"step": 600
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 7.29,
|
| 127 |
+
"grad_norm": 5.013394355773926,
|
| 128 |
+
"learning_rate": 9.166666666666666e-06,
|
| 129 |
+
"loss": 0.5745,
|
| 130 |
+
"step": 700
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 7.29,
|
| 134 |
+
"eval_accuracy": 0.9160953313744695,
|
| 135 |
+
"eval_f1_score": 0.8738968846084525,
|
| 136 |
+
"eval_loss": 0.6176044940948486,
|
| 137 |
+
"eval_precision": 0.8880590844686177,
|
| 138 |
+
"eval_recall": 0.8609982161968145,
|
| 139 |
+
"eval_runtime": 5.0009,
|
| 140 |
+
"eval_samples_per_second": 612.484,
|
| 141 |
+
"eval_steps_per_second": 9.598,
|
| 142 |
+
"step": 700
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.33,
|
| 146 |
+
"grad_norm": 4.954068183898926,
|
| 147 |
+
"learning_rate": 8.888888888888888e-06,
|
| 148 |
+
"loss": 0.5466,
|
| 149 |
+
"step": 800
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 8.33,
|
| 153 |
+
"eval_accuracy": 0.9180541952334312,
|
| 154 |
+
"eval_f1_score": 0.8730278944900508,
|
| 155 |
+
"eval_loss": 0.613418698310852,
|
| 156 |
+
"eval_precision": 0.8951285463624247,
|
| 157 |
+
"eval_recall": 0.8555146035099561,
|
| 158 |
+
"eval_runtime": 4.9947,
|
| 159 |
+
"eval_samples_per_second": 613.247,
|
| 160 |
+
"eval_steps_per_second": 9.61,
|
| 161 |
+
"step": 800
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.38,
|
| 165 |
+
"grad_norm": 6.174578666687012,
|
| 166 |
+
"learning_rate": 8.611111111111112e-06,
|
| 167 |
+
"loss": 0.5254,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 9.38,
|
| 172 |
+
"eval_accuracy": 0.9183806725432583,
|
| 173 |
+
"eval_f1_score": 0.8794225212758101,
|
| 174 |
+
"eval_loss": 0.6195659637451172,
|
| 175 |
+
"eval_precision": 0.898010742334885,
|
| 176 |
+
"eval_recall": 0.8634351648847699,
|
| 177 |
+
"eval_runtime": 4.9765,
|
| 178 |
+
"eval_samples_per_second": 615.487,
|
| 179 |
+
"eval_steps_per_second": 9.645,
|
| 180 |
+
"step": 900
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 10.42,
|
| 184 |
+
"grad_norm": 5.367175102233887,
|
| 185 |
+
"learning_rate": 8.333333333333334e-06,
|
| 186 |
+
"loss": 0.5059,
|
| 187 |
+
"step": 1000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 10.42,
|
| 191 |
+
"eval_accuracy": 0.9219719229513549,
|
| 192 |
+
"eval_f1_score": 0.8799283123746076,
|
| 193 |
+
"eval_loss": 0.6174536347389221,
|
| 194 |
+
"eval_precision": 0.8925411600987329,
|
| 195 |
+
"eval_recall": 0.8702271946202808,
|
| 196 |
+
"eval_runtime": 4.9764,
|
| 197 |
+
"eval_samples_per_second": 615.501,
|
| 198 |
+
"eval_steps_per_second": 9.645,
|
| 199 |
+
"step": 1000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 11.46,
|
| 203 |
+
"grad_norm": 6.586447238922119,
|
| 204 |
+
"learning_rate": 8.055555555555557e-06,
|
| 205 |
+
"loss": 0.4971,
|
| 206 |
+
"step": 1100
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 11.46,
|
| 210 |
+
"eval_accuracy": 0.9226248775710089,
|
| 211 |
+
"eval_f1_score": 0.8770727386787929,
|
| 212 |
+
"eval_loss": 0.6110312342643738,
|
| 213 |
+
"eval_precision": 0.89290929483548,
|
| 214 |
+
"eval_recall": 0.8638757728594787,
|
| 215 |
+
"eval_runtime": 4.9931,
|
| 216 |
+
"eval_samples_per_second": 613.443,
|
| 217 |
+
"eval_steps_per_second": 9.613,
|
| 218 |
+
"step": 1100
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 12.5,
|
| 222 |
+
"grad_norm": 3.8108513355255127,
|
| 223 |
+
"learning_rate": 7.77777777777778e-06,
|
| 224 |
+
"loss": 0.4872,
|
| 225 |
+
"step": 1200
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 12.5,
|
| 229 |
+
"eval_accuracy": 0.9229513548808358,
|
| 230 |
+
"eval_f1_score": 0.8833538508725888,
|
| 231 |
+
"eval_loss": 0.6191003918647766,
|
| 232 |
+
"eval_precision": 0.8914104564156814,
|
| 233 |
+
"eval_recall": 0.8774149435732262,
|
| 234 |
+
"eval_runtime": 4.9594,
|
| 235 |
+
"eval_samples_per_second": 617.612,
|
| 236 |
+
"eval_steps_per_second": 9.679,
|
| 237 |
+
"step": 1200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 13.54,
|
| 241 |
+
"grad_norm": 2.2206079959869385,
|
| 242 |
+
"learning_rate": 7.500000000000001e-06,
|
| 243 |
+
"loss": 0.4756,
|
| 244 |
+
"step": 1300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 13.54,
|
| 248 |
+
"eval_accuracy": 0.9242572641201436,
|
| 249 |
+
"eval_f1_score": 0.892919420274636,
|
| 250 |
+
"eval_loss": 0.6239792704582214,
|
| 251 |
+
"eval_precision": 0.8938258370033757,
|
| 252 |
+
"eval_recall": 0.8931296696096861,
|
| 253 |
+
"eval_runtime": 4.9692,
|
| 254 |
+
"eval_samples_per_second": 616.403,
|
| 255 |
+
"eval_steps_per_second": 9.66,
|
| 256 |
+
"step": 1300
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 14.58,
|
| 260 |
+
"grad_norm": 3.2590479850769043,
|
| 261 |
+
"learning_rate": 7.222222222222223e-06,
|
| 262 |
+
"loss": 0.4737,
|
| 263 |
+
"step": 1400
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 14.58,
|
| 267 |
+
"eval_accuracy": 0.92033953640222,
|
| 268 |
+
"eval_f1_score": 0.8793750306978456,
|
| 269 |
+
"eval_loss": 0.6246046423912048,
|
| 270 |
+
"eval_precision": 0.9038460171417751,
|
| 271 |
+
"eval_recall": 0.8582419368732399,
|
| 272 |
+
"eval_runtime": 5.0236,
|
| 273 |
+
"eval_samples_per_second": 609.724,
|
| 274 |
+
"eval_steps_per_second": 9.555,
|
| 275 |
+
"step": 1400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 15.62,
|
| 279 |
+
"grad_norm": 4.4676127433776855,
|
| 280 |
+
"learning_rate": 6.944444444444445e-06,
|
| 281 |
+
"loss": 0.4626,
|
| 282 |
+
"step": 1500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 15.62,
|
| 286 |
+
"eval_accuracy": 0.9249102187397976,
|
| 287 |
+
"eval_f1_score": 0.8889732013227639,
|
| 288 |
+
"eval_loss": 0.6267126798629761,
|
| 289 |
+
"eval_precision": 0.8955131575427445,
|
| 290 |
+
"eval_recall": 0.8842202196754521,
|
| 291 |
+
"eval_runtime": 5.0075,
|
| 292 |
+
"eval_samples_per_second": 611.678,
|
| 293 |
+
"eval_steps_per_second": 9.586,
|
| 294 |
+
"step": 1500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 16.67,
|
| 298 |
+
"grad_norm": 1.5825713872909546,
|
| 299 |
+
"learning_rate": 6.666666666666667e-06,
|
| 300 |
+
"loss": 0.4641,
|
| 301 |
+
"step": 1600
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 16.67,
|
| 305 |
+
"eval_accuracy": 0.9271955599085864,
|
| 306 |
+
"eval_f1_score": 0.8943646614708781,
|
| 307 |
+
"eval_loss": 0.6228047013282776,
|
| 308 |
+
"eval_precision": 0.9063376648453086,
|
| 309 |
+
"eval_recall": 0.8842502417202245,
|
| 310 |
+
"eval_runtime": 4.9845,
|
| 311 |
+
"eval_samples_per_second": 614.503,
|
| 312 |
+
"eval_steps_per_second": 9.63,
|
| 313 |
+
"step": 1600
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 17.71,
|
| 317 |
+
"grad_norm": 3.3820607662200928,
|
| 318 |
+
"learning_rate": 6.3888888888888885e-06,
|
| 319 |
+
"loss": 0.4562,
|
| 320 |
+
"step": 1700
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 17.71,
|
| 324 |
+
"eval_accuracy": 0.9278485145282402,
|
| 325 |
+
"eval_f1_score": 0.892329553955837,
|
| 326 |
+
"eval_loss": 0.6255541443824768,
|
| 327 |
+
"eval_precision": 0.9064974712840742,
|
| 328 |
+
"eval_recall": 0.8800313461131938,
|
| 329 |
+
"eval_runtime": 5.0301,
|
| 330 |
+
"eval_samples_per_second": 608.937,
|
| 331 |
+
"eval_steps_per_second": 9.543,
|
| 332 |
+
"step": 1700
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 18.75,
|
| 336 |
+
"grad_norm": 0.6632604598999023,
|
| 337 |
+
"learning_rate": 6.111111111111112e-06,
|
| 338 |
+
"loss": 0.4522,
|
| 339 |
+
"step": 1800
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 18.75,
|
| 343 |
+
"eval_accuracy": 0.9285014691478942,
|
| 344 |
+
"eval_f1_score": 0.8912735294131965,
|
| 345 |
+
"eval_loss": 0.6203188300132751,
|
| 346 |
+
"eval_precision": 0.8992701751319325,
|
| 347 |
+
"eval_recall": 0.8844935194666589,
|
| 348 |
+
"eval_runtime": 5.0719,
|
| 349 |
+
"eval_samples_per_second": 603.913,
|
| 350 |
+
"eval_steps_per_second": 9.464,
|
| 351 |
+
"step": 1800
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 19.79,
|
| 355 |
+
"grad_norm": 1.0484055280685425,
|
| 356 |
+
"learning_rate": 5.833333333333334e-06,
|
| 357 |
+
"loss": 0.4476,
|
| 358 |
+
"step": 1900
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 19.79,
|
| 362 |
+
"eval_accuracy": 0.9262161279791055,
|
| 363 |
+
"eval_f1_score": 0.8848887104133892,
|
| 364 |
+
"eval_loss": 0.6258328557014465,
|
| 365 |
+
"eval_precision": 0.9033804413868713,
|
| 366 |
+
"eval_recall": 0.8691664648450299,
|
| 367 |
+
"eval_runtime": 5.0588,
|
| 368 |
+
"eval_samples_per_second": 605.485,
|
| 369 |
+
"eval_steps_per_second": 9.488,
|
| 370 |
+
"step": 1900
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 20.83,
|
| 374 |
+
"grad_norm": 4.217446804046631,
|
| 375 |
+
"learning_rate": 5.555555555555557e-06,
|
| 376 |
+
"loss": 0.4474,
|
| 377 |
+
"step": 2000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 20.83,
|
| 381 |
+
"eval_accuracy": 0.9271955599085864,
|
| 382 |
+
"eval_f1_score": 0.8931632926835013,
|
| 383 |
+
"eval_loss": 0.6333132386207581,
|
| 384 |
+
"eval_precision": 0.9133975869931371,
|
| 385 |
+
"eval_recall": 0.8751018475655369,
|
| 386 |
+
"eval_runtime": 4.9412,
|
| 387 |
+
"eval_samples_per_second": 619.889,
|
| 388 |
+
"eval_steps_per_second": 9.714,
|
| 389 |
+
"step": 2000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 21.88,
|
| 393 |
+
"grad_norm": 2.5882654190063477,
|
| 394 |
+
"learning_rate": 5.2777777777777785e-06,
|
| 395 |
+
"loss": 0.4421,
|
| 396 |
+
"step": 2100
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 21.88,
|
| 400 |
+
"eval_accuracy": 0.9291544237675482,
|
| 401 |
+
"eval_f1_score": 0.8929419711474625,
|
| 402 |
+
"eval_loss": 0.6339655518531799,
|
| 403 |
+
"eval_precision": 0.8975107075254134,
|
| 404 |
+
"eval_recall": 0.8896075654961032,
|
| 405 |
+
"eval_runtime": 4.9672,
|
| 406 |
+
"eval_samples_per_second": 616.645,
|
| 407 |
+
"eval_steps_per_second": 9.663,
|
| 408 |
+
"step": 2100
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 21.88,
|
| 412 |
+
"step": 2100,
|
| 413 |
+
"total_flos": 4711306254898320.0,
|
| 414 |
+
"train_loss": 0.5936482783726283,
|
| 415 |
+
"train_runtime": 1811.9756,
|
| 416 |
+
"train_samples_per_second": 282.565,
|
| 417 |
+
"train_steps_per_second": 2.208
|
| 418 |
+
}
|
| 419 |
+
],
|
| 420 |
+
"logging_steps": 100,
|
| 421 |
+
"max_steps": 4000,
|
| 422 |
+
"num_input_tokens_seen": 0,
|
| 423 |
+
"num_train_epochs": 42,
|
| 424 |
+
"save_steps": 100,
|
| 425 |
+
"total_flos": 4711306254898320.0,
|
| 426 |
+
"train_batch_size": 64,
|
| 427 |
+
"trial_name": null,
|
| 428 |
+
"trial_params": null
|
| 429 |
+
}
|