End of training
Browse files- README.md +2 -2
- all_results.json +15 -0
- eval_results.json +9 -0
- predict_results.txt +0 -0
- train_results.json +9 -0
- trainer_state.json +619 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [vinai/phobert-base](https://huggingface.co/vinai/phobert-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- F1: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [vinai/phobert-base](https://huggingface.co/vinai/phobert-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.3677
|
| 22 |
+
- F1: 0.9408
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_f1": 0.9408151870463428,
|
| 4 |
+
"eval_loss": 0.36774319410324097,
|
| 5 |
+
"eval_runtime": 9.7511,
|
| 6 |
+
"eval_samples": 3582,
|
| 7 |
+
"eval_samples_per_second": 367.345,
|
| 8 |
+
"eval_steps_per_second": 5.743,
|
| 9 |
+
"total_flos": 6.517582470540288e+16,
|
| 10 |
+
"train_loss": 0.040758436324900574,
|
| 11 |
+
"train_runtime": 8835.4698,
|
| 12 |
+
"train_samples": 24771,
|
| 13 |
+
"train_samples_per_second": 112.143,
|
| 14 |
+
"train_steps_per_second": 1.757
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_f1": 0.9408151870463428,
|
| 4 |
+
"eval_loss": 0.36774319410324097,
|
| 5 |
+
"eval_runtime": 9.7511,
|
| 6 |
+
"eval_samples": 3582,
|
| 7 |
+
"eval_samples_per_second": 367.345,
|
| 8 |
+
"eval_steps_per_second": 5.743
|
| 9 |
+
}
|
predict_results.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 6.517582470540288e+16,
|
| 4 |
+
"train_loss": 0.040758436324900574,
|
| 5 |
+
"train_runtime": 8835.4698,
|
| 6 |
+
"train_samples": 24771,
|
| 7 |
+
"train_samples_per_second": 112.143,
|
| 8 |
+
"train_steps_per_second": 1.757
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,619 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9408151870463428,
|
| 3 |
+
"best_model_checkpoint": "/tmp/classification_phobertbase/checkpoint-5820",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 15520,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_f1": 0.9251814628699051,
|
| 14 |
+
"eval_loss": 0.21701952815055847,
|
| 15 |
+
"eval_runtime": 9.8556,
|
| 16 |
+
"eval_samples_per_second": 363.447,
|
| 17 |
+
"eval_steps_per_second": 5.682,
|
| 18 |
+
"step": 388
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.2886597938144329,
|
| 22 |
+
"grad_norm": 4.1991682052612305,
|
| 23 |
+
"learning_rate": 2.9033505154639176e-05,
|
| 24 |
+
"loss": 0.3098,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_f1": 0.9369067560022334,
|
| 30 |
+
"eval_loss": 0.2119385302066803,
|
| 31 |
+
"eval_runtime": 9.8764,
|
| 32 |
+
"eval_samples_per_second": 362.683,
|
| 33 |
+
"eval_steps_per_second": 5.67,
|
| 34 |
+
"step": 776
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.5773195876288657,
|
| 38 |
+
"grad_norm": 6.048534870147705,
|
| 39 |
+
"learning_rate": 2.806701030927835e-05,
|
| 40 |
+
"loss": 0.1763,
|
| 41 |
+
"step": 1000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"eval_f1": 0.932998324958124,
|
| 46 |
+
"eval_loss": 0.20233677327632904,
|
| 47 |
+
"eval_runtime": 9.8515,
|
| 48 |
+
"eval_samples_per_second": 363.598,
|
| 49 |
+
"eval_steps_per_second": 5.684,
|
| 50 |
+
"step": 1164
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.865979381443299,
|
| 54 |
+
"grad_norm": 3.9996299743652344,
|
| 55 |
+
"learning_rate": 2.7100515463917526e-05,
|
| 56 |
+
"loss": 0.1286,
|
| 57 |
+
"step": 1500
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"eval_f1": 0.9343941931881632,
|
| 62 |
+
"eval_loss": 0.26061517000198364,
|
| 63 |
+
"eval_runtime": 9.8544,
|
| 64 |
+
"eval_samples_per_second": 363.492,
|
| 65 |
+
"eval_steps_per_second": 5.683,
|
| 66 |
+
"step": 1552
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 5.0,
|
| 70 |
+
"eval_f1": 0.9391401451702959,
|
| 71 |
+
"eval_loss": 0.23778466880321503,
|
| 72 |
+
"eval_runtime": 9.8578,
|
| 73 |
+
"eval_samples_per_second": 363.368,
|
| 74 |
+
"eval_steps_per_second": 5.681,
|
| 75 |
+
"step": 1940
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.154639175257732,
|
| 79 |
+
"grad_norm": 3.624183416366577,
|
| 80 |
+
"learning_rate": 2.61340206185567e-05,
|
| 81 |
+
"loss": 0.0943,
|
| 82 |
+
"step": 2000
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 6.0,
|
| 86 |
+
"eval_f1": 0.9335566722501396,
|
| 87 |
+
"eval_loss": 0.2650683522224426,
|
| 88 |
+
"eval_runtime": 9.8594,
|
| 89 |
+
"eval_samples_per_second": 363.308,
|
| 90 |
+
"eval_steps_per_second": 5.68,
|
| 91 |
+
"step": 2328
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 6.443298969072165,
|
| 95 |
+
"grad_norm": 7.170408248901367,
|
| 96 |
+
"learning_rate": 2.5167525773195877e-05,
|
| 97 |
+
"loss": 0.0706,
|
| 98 |
+
"step": 2500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.0,
|
| 102 |
+
"eval_f1": 0.9366275823562256,
|
| 103 |
+
"eval_loss": 0.2985905408859253,
|
| 104 |
+
"eval_runtime": 9.8626,
|
| 105 |
+
"eval_samples_per_second": 363.191,
|
| 106 |
+
"eval_steps_per_second": 5.678,
|
| 107 |
+
"step": 2716
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 7.731958762886598,
|
| 111 |
+
"grad_norm": 2.5986905097961426,
|
| 112 |
+
"learning_rate": 2.4201030927835052e-05,
|
| 113 |
+
"loss": 0.0626,
|
| 114 |
+
"step": 3000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 8.0,
|
| 118 |
+
"eval_f1": 0.9304857621440537,
|
| 119 |
+
"eval_loss": 0.31677183508872986,
|
| 120 |
+
"eval_runtime": 9.879,
|
| 121 |
+
"eval_samples_per_second": 362.589,
|
| 122 |
+
"eval_steps_per_second": 5.669,
|
| 123 |
+
"step": 3104
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 9.0,
|
| 127 |
+
"eval_f1": 0.9357900614182021,
|
| 128 |
+
"eval_loss": 0.30201148986816406,
|
| 129 |
+
"eval_runtime": 9.8829,
|
| 130 |
+
"eval_samples_per_second": 362.444,
|
| 131 |
+
"eval_steps_per_second": 5.666,
|
| 132 |
+
"step": 3492
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.02061855670103,
|
| 136 |
+
"grad_norm": 0.1557622104883194,
|
| 137 |
+
"learning_rate": 2.3234536082474227e-05,
|
| 138 |
+
"loss": 0.0515,
|
| 139 |
+
"step": 3500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 10.0,
|
| 143 |
+
"eval_f1": 0.9366275823562256,
|
| 144 |
+
"eval_loss": 0.30621805787086487,
|
| 145 |
+
"eval_runtime": 9.8777,
|
| 146 |
+
"eval_samples_per_second": 362.634,
|
| 147 |
+
"eval_steps_per_second": 5.669,
|
| 148 |
+
"step": 3880
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 10.309278350515465,
|
| 152 |
+
"grad_norm": 0.21442590653896332,
|
| 153 |
+
"learning_rate": 2.2268041237113402e-05,
|
| 154 |
+
"loss": 0.0397,
|
| 155 |
+
"step": 4000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 11.0,
|
| 159 |
+
"eval_f1": 0.9343941931881632,
|
| 160 |
+
"eval_loss": 0.34871262311935425,
|
| 161 |
+
"eval_runtime": 9.8759,
|
| 162 |
+
"eval_samples_per_second": 362.7,
|
| 163 |
+
"eval_steps_per_second": 5.67,
|
| 164 |
+
"step": 4268
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 11.597938144329897,
|
| 168 |
+
"grad_norm": 13.442221641540527,
|
| 169 |
+
"learning_rate": 2.1301546391752577e-05,
|
| 170 |
+
"loss": 0.0337,
|
| 171 |
+
"step": 4500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 12.0,
|
| 175 |
+
"eval_f1": 0.9290898939140145,
|
| 176 |
+
"eval_loss": 0.4043188989162445,
|
| 177 |
+
"eval_runtime": 9.879,
|
| 178 |
+
"eval_samples_per_second": 362.586,
|
| 179 |
+
"eval_steps_per_second": 5.669,
|
| 180 |
+
"step": 4656
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 12.88659793814433,
|
| 184 |
+
"grad_norm": 0.3749821186065674,
|
| 185 |
+
"learning_rate": 2.0335051546391752e-05,
|
| 186 |
+
"loss": 0.031,
|
| 187 |
+
"step": 5000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 13.0,
|
| 191 |
+
"eval_f1": 0.9366275823562256,
|
| 192 |
+
"eval_loss": 0.37788987159729004,
|
| 193 |
+
"eval_runtime": 9.8649,
|
| 194 |
+
"eval_samples_per_second": 363.104,
|
| 195 |
+
"eval_steps_per_second": 5.677,
|
| 196 |
+
"step": 5044
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 14.0,
|
| 200 |
+
"eval_f1": 0.9293690675600224,
|
| 201 |
+
"eval_loss": 0.39344093203544617,
|
| 202 |
+
"eval_runtime": 9.8929,
|
| 203 |
+
"eval_samples_per_second": 362.077,
|
| 204 |
+
"eval_steps_per_second": 5.661,
|
| 205 |
+
"step": 5432
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 14.175257731958762,
|
| 209 |
+
"grad_norm": 0.08398671448230743,
|
| 210 |
+
"learning_rate": 1.936855670103093e-05,
|
| 211 |
+
"loss": 0.0266,
|
| 212 |
+
"step": 5500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 15.0,
|
| 216 |
+
"eval_f1": 0.9408151870463428,
|
| 217 |
+
"eval_loss": 0.36774319410324097,
|
| 218 |
+
"eval_runtime": 9.8647,
|
| 219 |
+
"eval_samples_per_second": 363.113,
|
| 220 |
+
"eval_steps_per_second": 5.677,
|
| 221 |
+
"step": 5820
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 15.463917525773196,
|
| 225 |
+
"grad_norm": 0.2722782790660858,
|
| 226 |
+
"learning_rate": 1.8402061855670106e-05,
|
| 227 |
+
"loss": 0.0246,
|
| 228 |
+
"step": 6000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 16.0,
|
| 232 |
+
"eval_f1": 0.9355108877721943,
|
| 233 |
+
"eval_loss": 0.3874046206474304,
|
| 234 |
+
"eval_runtime": 9.8701,
|
| 235 |
+
"eval_samples_per_second": 362.913,
|
| 236 |
+
"eval_steps_per_second": 5.674,
|
| 237 |
+
"step": 6208
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 16.75257731958763,
|
| 241 |
+
"grad_norm": 8.601401329040527,
|
| 242 |
+
"learning_rate": 1.743556701030928e-05,
|
| 243 |
+
"loss": 0.0222,
|
| 244 |
+
"step": 6500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 17.0,
|
| 248 |
+
"eval_f1": 0.9343941931881632,
|
| 249 |
+
"eval_loss": 0.42573627829551697,
|
| 250 |
+
"eval_runtime": 9.8744,
|
| 251 |
+
"eval_samples_per_second": 362.755,
|
| 252 |
+
"eval_steps_per_second": 5.671,
|
| 253 |
+
"step": 6596
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 18.0,
|
| 257 |
+
"eval_f1": 0.9369067560022334,
|
| 258 |
+
"eval_loss": 0.4371758997440338,
|
| 259 |
+
"eval_runtime": 9.8898,
|
| 260 |
+
"eval_samples_per_second": 362.193,
|
| 261 |
+
"eval_steps_per_second": 5.662,
|
| 262 |
+
"step": 6984
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 18.04123711340206,
|
| 266 |
+
"grad_norm": 0.05734695494174957,
|
| 267 |
+
"learning_rate": 1.6469072164948456e-05,
|
| 268 |
+
"loss": 0.022,
|
| 269 |
+
"step": 7000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 19.0,
|
| 273 |
+
"eval_f1": 0.9363484087102177,
|
| 274 |
+
"eval_loss": 0.44078370928764343,
|
| 275 |
+
"eval_runtime": 9.8828,
|
| 276 |
+
"eval_samples_per_second": 362.448,
|
| 277 |
+
"eval_steps_per_second": 5.666,
|
| 278 |
+
"step": 7372
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 19.329896907216494,
|
| 282 |
+
"grad_norm": 0.19072239100933075,
|
| 283 |
+
"learning_rate": 1.550257731958763e-05,
|
| 284 |
+
"loss": 0.0176,
|
| 285 |
+
"step": 7500
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 20.0,
|
| 289 |
+
"eval_f1": 0.9357900614182021,
|
| 290 |
+
"eval_loss": 0.4601401090621948,
|
| 291 |
+
"eval_runtime": 9.8712,
|
| 292 |
+
"eval_samples_per_second": 362.874,
|
| 293 |
+
"eval_steps_per_second": 5.673,
|
| 294 |
+
"step": 7760
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 20.61855670103093,
|
| 298 |
+
"grad_norm": 0.14792193472385406,
|
| 299 |
+
"learning_rate": 1.4536082474226805e-05,
|
| 300 |
+
"loss": 0.0142,
|
| 301 |
+
"step": 8000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 21.0,
|
| 305 |
+
"eval_f1": 0.9360692350642099,
|
| 306 |
+
"eval_loss": 0.45033514499664307,
|
| 307 |
+
"eval_runtime": 9.8715,
|
| 308 |
+
"eval_samples_per_second": 362.861,
|
| 309 |
+
"eval_steps_per_second": 5.673,
|
| 310 |
+
"step": 8148
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 21.90721649484536,
|
| 314 |
+
"grad_norm": 0.03246806561946869,
|
| 315 |
+
"learning_rate": 1.356958762886598e-05,
|
| 316 |
+
"loss": 0.0134,
|
| 317 |
+
"step": 8500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 22.0,
|
| 321 |
+
"eval_f1": 0.9366275823562256,
|
| 322 |
+
"eval_loss": 0.4835437834262848,
|
| 323 |
+
"eval_runtime": 9.8869,
|
| 324 |
+
"eval_samples_per_second": 362.298,
|
| 325 |
+
"eval_steps_per_second": 5.664,
|
| 326 |
+
"step": 8536
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 23.0,
|
| 330 |
+
"eval_f1": 0.9391401451702959,
|
| 331 |
+
"eval_loss": 0.45941534638404846,
|
| 332 |
+
"eval_runtime": 9.8785,
|
| 333 |
+
"eval_samples_per_second": 362.605,
|
| 334 |
+
"eval_steps_per_second": 5.669,
|
| 335 |
+
"step": 8924
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 23.195876288659793,
|
| 339 |
+
"grad_norm": 0.08111796528100967,
|
| 340 |
+
"learning_rate": 1.2603092783505155e-05,
|
| 341 |
+
"loss": 0.0126,
|
| 342 |
+
"step": 9000
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 24.0,
|
| 346 |
+
"eval_f1": 0.9366275823562256,
|
| 347 |
+
"eval_loss": 0.48089537024497986,
|
| 348 |
+
"eval_runtime": 9.8641,
|
| 349 |
+
"eval_samples_per_second": 363.135,
|
| 350 |
+
"eval_steps_per_second": 5.677,
|
| 351 |
+
"step": 9312
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 24.484536082474225,
|
| 355 |
+
"grad_norm": 0.06357069313526154,
|
| 356 |
+
"learning_rate": 1.163659793814433e-05,
|
| 357 |
+
"loss": 0.0109,
|
| 358 |
+
"step": 9500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 25.0,
|
| 362 |
+
"eval_f1": 0.9369067560022334,
|
| 363 |
+
"eval_loss": 0.48593300580978394,
|
| 364 |
+
"eval_runtime": 9.8681,
|
| 365 |
+
"eval_samples_per_second": 362.988,
|
| 366 |
+
"eval_steps_per_second": 5.675,
|
| 367 |
+
"step": 9700
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 25.77319587628866,
|
| 371 |
+
"grad_norm": 0.006769911386072636,
|
| 372 |
+
"learning_rate": 1.0670103092783506e-05,
|
| 373 |
+
"loss": 0.012,
|
| 374 |
+
"step": 10000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 26.0,
|
| 378 |
+
"eval_f1": 0.9385817978782803,
|
| 379 |
+
"eval_loss": 0.48239514231681824,
|
| 380 |
+
"eval_runtime": 9.8778,
|
| 381 |
+
"eval_samples_per_second": 362.631,
|
| 382 |
+
"eval_steps_per_second": 5.669,
|
| 383 |
+
"step": 10088
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 27.0,
|
| 387 |
+
"eval_f1": 0.9360692350642099,
|
| 388 |
+
"eval_loss": 0.5067496299743652,
|
| 389 |
+
"eval_runtime": 9.8844,
|
| 390 |
+
"eval_samples_per_second": 362.39,
|
| 391 |
+
"eval_steps_per_second": 5.666,
|
| 392 |
+
"step": 10476
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 27.061855670103093,
|
| 396 |
+
"grad_norm": 0.037592578679323196,
|
| 397 |
+
"learning_rate": 9.703608247422681e-06,
|
| 398 |
+
"loss": 0.0101,
|
| 399 |
+
"step": 10500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 28.0,
|
| 403 |
+
"eval_f1": 0.937465103294249,
|
| 404 |
+
"eval_loss": 0.48702046275138855,
|
| 405 |
+
"eval_runtime": 9.8977,
|
| 406 |
+
"eval_samples_per_second": 361.901,
|
| 407 |
+
"eval_steps_per_second": 5.658,
|
| 408 |
+
"step": 10864
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 28.350515463917525,
|
| 412 |
+
"grad_norm": 0.12630492448806763,
|
| 413 |
+
"learning_rate": 8.737113402061856e-06,
|
| 414 |
+
"loss": 0.0102,
|
| 415 |
+
"step": 11000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 29.0,
|
| 419 |
+
"eval_f1": 0.9355108877721943,
|
| 420 |
+
"eval_loss": 0.5301576256752014,
|
| 421 |
+
"eval_runtime": 9.8877,
|
| 422 |
+
"eval_samples_per_second": 362.268,
|
| 423 |
+
"eval_steps_per_second": 5.664,
|
| 424 |
+
"step": 11252
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 29.63917525773196,
|
| 428 |
+
"grad_norm": 0.2997748851776123,
|
| 429 |
+
"learning_rate": 7.770618556701031e-06,
|
| 430 |
+
"loss": 0.0088,
|
| 431 |
+
"step": 11500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 30.0,
|
| 435 |
+
"eval_f1": 0.9366275823562256,
|
| 436 |
+
"eval_loss": 0.49529990553855896,
|
| 437 |
+
"eval_runtime": 9.8876,
|
| 438 |
+
"eval_samples_per_second": 362.27,
|
| 439 |
+
"eval_steps_per_second": 5.664,
|
| 440 |
+
"step": 11640
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 30.927835051546392,
|
| 444 |
+
"grad_norm": 0.33649611473083496,
|
| 445 |
+
"learning_rate": 6.8041237113402065e-06,
|
| 446 |
+
"loss": 0.0093,
|
| 447 |
+
"step": 12000
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 31.0,
|
| 451 |
+
"eval_f1": 0.9360692350642099,
|
| 452 |
+
"eval_loss": 0.4913821220397949,
|
| 453 |
+
"eval_runtime": 9.8622,
|
| 454 |
+
"eval_samples_per_second": 363.206,
|
| 455 |
+
"eval_steps_per_second": 5.678,
|
| 456 |
+
"step": 12028
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 32.0,
|
| 460 |
+
"eval_f1": 0.9388609715242882,
|
| 461 |
+
"eval_loss": 0.5014358162879944,
|
| 462 |
+
"eval_runtime": 9.8775,
|
| 463 |
+
"eval_samples_per_second": 362.641,
|
| 464 |
+
"eval_steps_per_second": 5.669,
|
| 465 |
+
"step": 12416
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 32.21649484536083,
|
| 469 |
+
"grad_norm": 0.011193674989044666,
|
| 470 |
+
"learning_rate": 5.837628865979382e-06,
|
| 471 |
+
"loss": 0.0084,
|
| 472 |
+
"step": 12500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 33.0,
|
| 476 |
+
"eval_f1": 0.9383026242322725,
|
| 477 |
+
"eval_loss": 0.5026176571846008,
|
| 478 |
+
"eval_runtime": 9.8893,
|
| 479 |
+
"eval_samples_per_second": 362.21,
|
| 480 |
+
"eval_steps_per_second": 5.663,
|
| 481 |
+
"step": 12804
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 33.50515463917526,
|
| 485 |
+
"grad_norm": 0.26669424772262573,
|
| 486 |
+
"learning_rate": 4.871134020618557e-06,
|
| 487 |
+
"loss": 0.0078,
|
| 488 |
+
"step": 13000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 34.0,
|
| 492 |
+
"eval_f1": 0.9380234505862647,
|
| 493 |
+
"eval_loss": 0.5043131709098816,
|
| 494 |
+
"eval_runtime": 9.8862,
|
| 495 |
+
"eval_samples_per_second": 362.323,
|
| 496 |
+
"eval_steps_per_second": 5.664,
|
| 497 |
+
"step": 13192
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 34.79381443298969,
|
| 501 |
+
"grad_norm": 0.9884344339370728,
|
| 502 |
+
"learning_rate": 3.904639175257732e-06,
|
| 503 |
+
"loss": 0.0075,
|
| 504 |
+
"step": 13500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 35.0,
|
| 508 |
+
"eval_f1": 0.9377442769402569,
|
| 509 |
+
"eval_loss": 0.5035226941108704,
|
| 510 |
+
"eval_runtime": 9.8894,
|
| 511 |
+
"eval_samples_per_second": 362.204,
|
| 512 |
+
"eval_steps_per_second": 5.663,
|
| 513 |
+
"step": 13580
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 36.0,
|
| 517 |
+
"eval_f1": 0.9377442769402569,
|
| 518 |
+
"eval_loss": 0.5007095336914062,
|
| 519 |
+
"eval_runtime": 9.9011,
|
| 520 |
+
"eval_samples_per_second": 361.777,
|
| 521 |
+
"eval_steps_per_second": 5.656,
|
| 522 |
+
"step": 13968
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 36.08247422680412,
|
| 526 |
+
"grad_norm": 0.02359125204384327,
|
| 527 |
+
"learning_rate": 2.938144329896907e-06,
|
| 528 |
+
"loss": 0.0077,
|
| 529 |
+
"step": 14000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 37.0,
|
| 533 |
+
"eval_f1": 0.9377442769402569,
|
| 534 |
+
"eval_loss": 0.5102269053459167,
|
| 535 |
+
"eval_runtime": 9.8918,
|
| 536 |
+
"eval_samples_per_second": 362.118,
|
| 537 |
+
"eval_steps_per_second": 5.661,
|
| 538 |
+
"step": 14356
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 37.371134020618555,
|
| 542 |
+
"grad_norm": 0.00852921698242426,
|
| 543 |
+
"learning_rate": 1.9716494845360827e-06,
|
| 544 |
+
"loss": 0.0076,
|
| 545 |
+
"step": 14500
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 38.0,
|
| 549 |
+
"eval_f1": 0.9391401451702959,
|
| 550 |
+
"eval_loss": 0.5068992376327515,
|
| 551 |
+
"eval_runtime": 9.8982,
|
| 552 |
+
"eval_samples_per_second": 361.883,
|
| 553 |
+
"eval_steps_per_second": 5.658,
|
| 554 |
+
"step": 14744
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 38.65979381443299,
|
| 558 |
+
"grad_norm": 0.01492872554808855,
|
| 559 |
+
"learning_rate": 1.0051546391752577e-06,
|
| 560 |
+
"loss": 0.0059,
|
| 561 |
+
"step": 15000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 39.0,
|
| 565 |
+
"eval_f1": 0.9385817978782803,
|
| 566 |
+
"eval_loss": 0.5104745030403137,
|
| 567 |
+
"eval_runtime": 9.9017,
|
| 568 |
+
"eval_samples_per_second": 361.758,
|
| 569 |
+
"eval_steps_per_second": 5.656,
|
| 570 |
+
"step": 15132
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 39.94845360824742,
|
| 574 |
+
"grad_norm": 0.013034285046160221,
|
| 575 |
+
"learning_rate": 3.865979381443299e-08,
|
| 576 |
+
"loss": 0.0071,
|
| 577 |
+
"step": 15500
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 40.0,
|
| 581 |
+
"eval_f1": 0.9383026242322725,
|
| 582 |
+
"eval_loss": 0.5110998749732971,
|
| 583 |
+
"eval_runtime": 9.8692,
|
| 584 |
+
"eval_samples_per_second": 362.949,
|
| 585 |
+
"eval_steps_per_second": 5.674,
|
| 586 |
+
"step": 15520
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 40.0,
|
| 590 |
+
"step": 15520,
|
| 591 |
+
"total_flos": 6.517582470540288e+16,
|
| 592 |
+
"train_loss": 0.040758436324900574,
|
| 593 |
+
"train_runtime": 8835.4698,
|
| 594 |
+
"train_samples_per_second": 112.143,
|
| 595 |
+
"train_steps_per_second": 1.757
|
| 596 |
+
}
|
| 597 |
+
],
|
| 598 |
+
"logging_steps": 500,
|
| 599 |
+
"max_steps": 15520,
|
| 600 |
+
"num_input_tokens_seen": 0,
|
| 601 |
+
"num_train_epochs": 40,
|
| 602 |
+
"save_steps": 500,
|
| 603 |
+
"stateful_callbacks": {
|
| 604 |
+
"TrainerControl": {
|
| 605 |
+
"args": {
|
| 606 |
+
"should_epoch_stop": false,
|
| 607 |
+
"should_evaluate": false,
|
| 608 |
+
"should_log": false,
|
| 609 |
+
"should_save": true,
|
| 610 |
+
"should_training_stop": true
|
| 611 |
+
},
|
| 612 |
+
"attributes": {}
|
| 613 |
+
}
|
| 614 |
+
},
|
| 615 |
+
"total_flos": 6.517582470540288e+16,
|
| 616 |
+
"train_batch_size": 64,
|
| 617 |
+
"trial_name": null,
|
| 618 |
+
"trial_params": null
|
| 619 |
+
}
|