Commit
•
84402d3
1
Parent(s):
d6a34bb
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 +600 -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: 0.2873
|
| 22 |
+
- Accuracy: 0.9323
|
| 23 |
+
- F1 Score: 0.9262
|
| 24 |
+
- Recall: 0.9217
|
| 25 |
+
- Precision: 0.9320
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.08695652173913,
|
| 3 |
+
"eval_accuracy": 0.9323185648273987,
|
| 4 |
+
"eval_f1_score": 0.9261695691084951,
|
| 5 |
+
"eval_loss": 0.2873239815235138,
|
| 6 |
+
"eval_precision": 0.9319834922740233,
|
| 7 |
+
"eval_recall": 0.9216726996777184,
|
| 8 |
+
"eval_runtime": 6.6463,
|
| 9 |
+
"eval_samples_per_second": 553.538,
|
| 10 |
+
"eval_steps_per_second": 8.727,
|
| 11 |
+
"total_flos": 6579999363349350.0,
|
| 12 |
+
"train_loss": 0.31640464369455973,
|
| 13 |
+
"train_runtime": 3238.6183,
|
| 14 |
+
"train_samples_per_second": 158.092,
|
| 15 |
+
"train_steps_per_second": 1.235
|
| 16 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.08695652173913,
|
| 3 |
+
"eval_accuracy": 0.9323185648273987,
|
| 4 |
+
"eval_f1_score": 0.9261695691084951,
|
| 5 |
+
"eval_loss": 0.2873239815235138,
|
| 6 |
+
"eval_precision": 0.9319834922740233,
|
| 7 |
+
"eval_recall": 0.9216726996777184,
|
| 8 |
+
"eval_runtime": 6.6463,
|
| 9 |
+
"eval_samples_per_second": 553.538,
|
| 10 |
+
"eval_steps_per_second": 8.727
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.08695652173913,
|
| 3 |
+
"total_flos": 6579999363349350.0,
|
| 4 |
+
"train_loss": 0.31640464369455973,
|
| 5 |
+
"train_runtime": 3238.6183,
|
| 6 |
+
"train_samples_per_second": 158.092,
|
| 7 |
+
"train_steps_per_second": 1.235
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,600 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9261695691084951,
|
| 3 |
+
"best_model_checkpoint": "cls_comment-phobert-base-v2-v3.2.1/checkpoint-2000",
|
| 4 |
+
"epoch": 26.08695652173913,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 3000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8695652173913043,
|
| 13 |
+
"grad_norm": 2.4986989498138428,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 1.8947,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.8695652173913043,
|
| 20 |
+
"eval_accuracy": 0.4001087251970644,
|
| 21 |
+
"eval_f1_score": 0.08320411950694513,
|
| 22 |
+
"eval_loss": 1.68748140335083,
|
| 23 |
+
"eval_precision": 0.14637409036074248,
|
| 24 |
+
"eval_recall": 0.14367816091954025,
|
| 25 |
+
"eval_runtime": 6.6876,
|
| 26 |
+
"eval_samples_per_second": 550.125,
|
| 27 |
+
"eval_steps_per_second": 8.673,
|
| 28 |
+
"step": 100
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.7391304347826086,
|
| 32 |
+
"grad_norm": 1.596021294593811,
|
| 33 |
+
"learning_rate": 5e-06,
|
| 34 |
+
"loss": 1.5395,
|
| 35 |
+
"step": 200
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.7391304347826086,
|
| 39 |
+
"eval_accuracy": 0.5849415602065778,
|
| 40 |
+
"eval_f1_score": 0.2355807809182458,
|
| 41 |
+
"eval_loss": 1.2897096872329712,
|
| 42 |
+
"eval_precision": 0.27516139357553443,
|
| 43 |
+
"eval_recall": 0.26320585050663253,
|
| 44 |
+
"eval_runtime": 6.7112,
|
| 45 |
+
"eval_samples_per_second": 548.189,
|
| 46 |
+
"eval_steps_per_second": 8.642,
|
| 47 |
+
"step": 200
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.608695652173913,
|
| 51 |
+
"grad_norm": 5.161496162414551,
|
| 52 |
+
"learning_rate": 7.500000000000001e-06,
|
| 53 |
+
"loss": 1.1205,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.608695652173913,
|
| 58 |
+
"eval_accuracy": 0.7999456374014677,
|
| 59 |
+
"eval_f1_score": 0.5833178086765388,
|
| 60 |
+
"eval_loss": 0.8468331098556519,
|
| 61 |
+
"eval_precision": 0.5889764394952819,
|
| 62 |
+
"eval_recall": 0.5810488238671468,
|
| 63 |
+
"eval_runtime": 6.6686,
|
| 64 |
+
"eval_samples_per_second": 551.694,
|
| 65 |
+
"eval_steps_per_second": 8.698,
|
| 66 |
+
"step": 300
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 3.4782608695652173,
|
| 70 |
+
"grad_norm": 5.227330207824707,
|
| 71 |
+
"learning_rate": 1e-05,
|
| 72 |
+
"loss": 0.82,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 3.4782608695652173,
|
| 77 |
+
"eval_accuracy": 0.8369122044033704,
|
| 78 |
+
"eval_f1_score": 0.6179371343772609,
|
| 79 |
+
"eval_loss": 0.6537477374076843,
|
| 80 |
+
"eval_precision": 0.6062100200393906,
|
| 81 |
+
"eval_recall": 0.6355302315827523,
|
| 82 |
+
"eval_runtime": 6.728,
|
| 83 |
+
"eval_samples_per_second": 546.823,
|
| 84 |
+
"eval_steps_per_second": 8.621,
|
| 85 |
+
"step": 400
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4.3478260869565215,
|
| 89 |
+
"grad_norm": 5.6816534996032715,
|
| 90 |
+
"learning_rate": 9.722222222222223e-06,
|
| 91 |
+
"loss": 0.6232,
|
| 92 |
+
"step": 500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.3478260869565215,
|
| 96 |
+
"eval_accuracy": 0.8537646099483556,
|
| 97 |
+
"eval_f1_score": 0.633743239294036,
|
| 98 |
+
"eval_loss": 0.537100613117218,
|
| 99 |
+
"eval_precision": 0.7525070200257705,
|
| 100 |
+
"eval_recall": 0.6518017678843925,
|
| 101 |
+
"eval_runtime": 6.7932,
|
| 102 |
+
"eval_samples_per_second": 541.573,
|
| 103 |
+
"eval_steps_per_second": 8.538,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 5.217391304347826,
|
| 108 |
+
"grad_norm": 5.096814155578613,
|
| 109 |
+
"learning_rate": 9.444444444444445e-06,
|
| 110 |
+
"loss": 0.5148,
|
| 111 |
+
"step": 600
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 5.217391304347826,
|
| 115 |
+
"eval_accuracy": 0.872791519434629,
|
| 116 |
+
"eval_f1_score": 0.7299293979398146,
|
| 117 |
+
"eval_loss": 0.46505650877952576,
|
| 118 |
+
"eval_precision": 0.7548552896750885,
|
| 119 |
+
"eval_recall": 0.7210618976649555,
|
| 120 |
+
"eval_runtime": 6.7028,
|
| 121 |
+
"eval_samples_per_second": 548.875,
|
| 122 |
+
"eval_steps_per_second": 8.653,
|
| 123 |
+
"step": 600
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 6.086956521739131,
|
| 127 |
+
"grad_norm": 5.458530902862549,
|
| 128 |
+
"learning_rate": 9.166666666666666e-06,
|
| 129 |
+
"loss": 0.4204,
|
| 130 |
+
"step": 700
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 6.086956521739131,
|
| 134 |
+
"eval_accuracy": 0.8869257950530035,
|
| 135 |
+
"eval_f1_score": 0.7654329783869755,
|
| 136 |
+
"eval_loss": 0.40097591280937195,
|
| 137 |
+
"eval_precision": 0.8914471413846636,
|
| 138 |
+
"eval_recall": 0.7712133932759179,
|
| 139 |
+
"eval_runtime": 6.7443,
|
| 140 |
+
"eval_samples_per_second": 545.497,
|
| 141 |
+
"eval_steps_per_second": 8.6,
|
| 142 |
+
"step": 700
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 6.956521739130435,
|
| 146 |
+
"grad_norm": 6.144416809082031,
|
| 147 |
+
"learning_rate": 8.888888888888888e-06,
|
| 148 |
+
"loss": 0.3421,
|
| 149 |
+
"step": 800
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 6.956521739130435,
|
| 153 |
+
"eval_accuracy": 0.9051372655612938,
|
| 154 |
+
"eval_f1_score": 0.8713582894968701,
|
| 155 |
+
"eval_loss": 0.3648029565811157,
|
| 156 |
+
"eval_precision": 0.8940734807154502,
|
| 157 |
+
"eval_recall": 0.8588405388993653,
|
| 158 |
+
"eval_runtime": 6.7352,
|
| 159 |
+
"eval_samples_per_second": 546.232,
|
| 160 |
+
"eval_steps_per_second": 8.611,
|
| 161 |
+
"step": 800
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 7.826086956521739,
|
| 165 |
+
"grad_norm": 9.907292366027832,
|
| 166 |
+
"learning_rate": 8.611111111111112e-06,
|
| 167 |
+
"loss": 0.2841,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.826086956521739,
|
| 172 |
+
"eval_accuracy": 0.9181842892090242,
|
| 173 |
+
"eval_f1_score": 0.9006880118200489,
|
| 174 |
+
"eval_loss": 0.3239505887031555,
|
| 175 |
+
"eval_precision": 0.8978148514278343,
|
| 176 |
+
"eval_recall": 0.9038252102525616,
|
| 177 |
+
"eval_runtime": 6.7315,
|
| 178 |
+
"eval_samples_per_second": 546.532,
|
| 179 |
+
"eval_steps_per_second": 8.616,
|
| 180 |
+
"step": 900
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 8.695652173913043,
|
| 184 |
+
"grad_norm": 6.941843032836914,
|
| 185 |
+
"learning_rate": 8.333333333333334e-06,
|
| 186 |
+
"loss": 0.2319,
|
| 187 |
+
"step": 1000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 8.695652173913043,
|
| 191 |
+
"eval_accuracy": 0.9203587931503125,
|
| 192 |
+
"eval_f1_score": 0.9060754755748909,
|
| 193 |
+
"eval_loss": 0.3025033473968506,
|
| 194 |
+
"eval_precision": 0.9175362378163865,
|
| 195 |
+
"eval_recall": 0.8975903509513042,
|
| 196 |
+
"eval_runtime": 6.6908,
|
| 197 |
+
"eval_samples_per_second": 549.861,
|
| 198 |
+
"eval_steps_per_second": 8.669,
|
| 199 |
+
"step": 1000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 9.565217391304348,
|
| 203 |
+
"grad_norm": 4.255012035369873,
|
| 204 |
+
"learning_rate": 8.055555555555557e-06,
|
| 205 |
+
"loss": 0.205,
|
| 206 |
+
"step": 1100
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 9.565217391304348,
|
| 210 |
+
"eval_accuracy": 0.9209024191356346,
|
| 211 |
+
"eval_f1_score": 0.9098640550303895,
|
| 212 |
+
"eval_loss": 0.29862046241760254,
|
| 213 |
+
"eval_precision": 0.9123097696068861,
|
| 214 |
+
"eval_recall": 0.9086287269577242,
|
| 215 |
+
"eval_runtime": 6.7134,
|
| 216 |
+
"eval_samples_per_second": 548.01,
|
| 217 |
+
"eval_steps_per_second": 8.639,
|
| 218 |
+
"step": 1100
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 10.434782608695652,
|
| 222 |
+
"grad_norm": 5.848569393157959,
|
| 223 |
+
"learning_rate": 7.77777777777778e-06,
|
| 224 |
+
"loss": 0.1783,
|
| 225 |
+
"step": 1200
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 10.434782608695652,
|
| 229 |
+
"eval_accuracy": 0.9206306061429737,
|
| 230 |
+
"eval_f1_score": 0.9104384776037051,
|
| 231 |
+
"eval_loss": 0.3047122657299042,
|
| 232 |
+
"eval_precision": 0.9024848857165658,
|
| 233 |
+
"eval_recall": 0.9207396220750284,
|
| 234 |
+
"eval_runtime": 6.6561,
|
| 235 |
+
"eval_samples_per_second": 552.726,
|
| 236 |
+
"eval_steps_per_second": 8.714,
|
| 237 |
+
"step": 1200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 11.304347826086957,
|
| 241 |
+
"grad_norm": 7.340043544769287,
|
| 242 |
+
"learning_rate": 7.500000000000001e-06,
|
| 243 |
+
"loss": 0.1587,
|
| 244 |
+
"step": 1300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 11.304347826086957,
|
| 248 |
+
"eval_accuracy": 0.9296004349007883,
|
| 249 |
+
"eval_f1_score": 0.9202832724978299,
|
| 250 |
+
"eval_loss": 0.2757803201675415,
|
| 251 |
+
"eval_precision": 0.9233347498988893,
|
| 252 |
+
"eval_recall": 0.917658614989255,
|
| 253 |
+
"eval_runtime": 6.6787,
|
| 254 |
+
"eval_samples_per_second": 550.859,
|
| 255 |
+
"eval_steps_per_second": 8.684,
|
| 256 |
+
"step": 1300
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 12.173913043478262,
|
| 260 |
+
"grad_norm": 5.315700054168701,
|
| 261 |
+
"learning_rate": 7.222222222222223e-06,
|
| 262 |
+
"loss": 0.1286,
|
| 263 |
+
"step": 1400
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 12.173913043478262,
|
| 267 |
+
"eval_accuracy": 0.9266104919815167,
|
| 268 |
+
"eval_f1_score": 0.9144278995332229,
|
| 269 |
+
"eval_loss": 0.29267847537994385,
|
| 270 |
+
"eval_precision": 0.9100638576136009,
|
| 271 |
+
"eval_recall": 0.9198715425139269,
|
| 272 |
+
"eval_runtime": 6.7676,
|
| 273 |
+
"eval_samples_per_second": 543.619,
|
| 274 |
+
"eval_steps_per_second": 8.57,
|
| 275 |
+
"step": 1400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 13.043478260869565,
|
| 279 |
+
"grad_norm": 5.173799514770508,
|
| 280 |
+
"learning_rate": 6.944444444444445e-06,
|
| 281 |
+
"loss": 0.1221,
|
| 282 |
+
"step": 1500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 13.043478260869565,
|
| 286 |
+
"eval_accuracy": 0.9317749388420766,
|
| 287 |
+
"eval_f1_score": 0.9245023460604546,
|
| 288 |
+
"eval_loss": 0.28211963176727295,
|
| 289 |
+
"eval_precision": 0.9309417300478454,
|
| 290 |
+
"eval_recall": 0.9193579289135766,
|
| 291 |
+
"eval_runtime": 6.7091,
|
| 292 |
+
"eval_samples_per_second": 548.359,
|
| 293 |
+
"eval_steps_per_second": 8.645,
|
| 294 |
+
"step": 1500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.91304347826087,
|
| 298 |
+
"grad_norm": 8.639619827270508,
|
| 299 |
+
"learning_rate": 6.666666666666667e-06,
|
| 300 |
+
"loss": 0.1087,
|
| 301 |
+
"step": 1600
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 13.91304347826087,
|
| 305 |
+
"eval_accuracy": 0.9293286219081273,
|
| 306 |
+
"eval_f1_score": 0.9159607873769989,
|
| 307 |
+
"eval_loss": 0.27890825271606445,
|
| 308 |
+
"eval_precision": 0.9090390134661626,
|
| 309 |
+
"eval_recall": 0.9236924050215896,
|
| 310 |
+
"eval_runtime": 6.7017,
|
| 311 |
+
"eval_samples_per_second": 548.966,
|
| 312 |
+
"eval_steps_per_second": 8.655,
|
| 313 |
+
"step": 1600
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 14.782608695652174,
|
| 317 |
+
"grad_norm": 6.424872398376465,
|
| 318 |
+
"learning_rate": 6.3888888888888885e-06,
|
| 319 |
+
"loss": 0.0982,
|
| 320 |
+
"step": 1700
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 14.782608695652174,
|
| 324 |
+
"eval_accuracy": 0.9290568089154662,
|
| 325 |
+
"eval_f1_score": 0.9196461825352048,
|
| 326 |
+
"eval_loss": 0.2833573520183563,
|
| 327 |
+
"eval_precision": 0.9187836187318232,
|
| 328 |
+
"eval_recall": 0.9213402050339831,
|
| 329 |
+
"eval_runtime": 6.7096,
|
| 330 |
+
"eval_samples_per_second": 548.32,
|
| 331 |
+
"eval_steps_per_second": 8.644,
|
| 332 |
+
"step": 1700
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 15.652173913043478,
|
| 336 |
+
"grad_norm": 4.618613243103027,
|
| 337 |
+
"learning_rate": 6.111111111111112e-06,
|
| 338 |
+
"loss": 0.089,
|
| 339 |
+
"step": 1800
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 15.652173913043478,
|
| 343 |
+
"eval_accuracy": 0.9298722478934494,
|
| 344 |
+
"eval_f1_score": 0.9202166850732406,
|
| 345 |
+
"eval_loss": 0.28280356526374817,
|
| 346 |
+
"eval_precision": 0.9151663252588741,
|
| 347 |
+
"eval_recall": 0.9260674008256092,
|
| 348 |
+
"eval_runtime": 6.7345,
|
| 349 |
+
"eval_samples_per_second": 546.292,
|
| 350 |
+
"eval_steps_per_second": 8.612,
|
| 351 |
+
"step": 1800
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 16.52173913043478,
|
| 355 |
+
"grad_norm": 1.9568698406219482,
|
| 356 |
+
"learning_rate": 5.833333333333334e-06,
|
| 357 |
+
"loss": 0.0795,
|
| 358 |
+
"step": 1900
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 16.52173913043478,
|
| 362 |
+
"eval_accuracy": 0.9331340038053819,
|
| 363 |
+
"eval_f1_score": 0.9244095368032713,
|
| 364 |
+
"eval_loss": 0.273701936006546,
|
| 365 |
+
"eval_precision": 0.925343846727414,
|
| 366 |
+
"eval_recall": 0.9238732382441093,
|
| 367 |
+
"eval_runtime": 6.7425,
|
| 368 |
+
"eval_samples_per_second": 545.641,
|
| 369 |
+
"eval_steps_per_second": 8.602,
|
| 370 |
+
"step": 1900
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 17.391304347826086,
|
| 374 |
+
"grad_norm": 2.161759614944458,
|
| 375 |
+
"learning_rate": 5.555555555555557e-06,
|
| 376 |
+
"loss": 0.0684,
|
| 377 |
+
"step": 2000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 17.391304347826086,
|
| 381 |
+
"eval_accuracy": 0.9323185648273987,
|
| 382 |
+
"eval_f1_score": 0.9261695691084951,
|
| 383 |
+
"eval_loss": 0.2873239815235138,
|
| 384 |
+
"eval_precision": 0.9319834922740233,
|
| 385 |
+
"eval_recall": 0.9216726996777184,
|
| 386 |
+
"eval_runtime": 6.7577,
|
| 387 |
+
"eval_samples_per_second": 544.415,
|
| 388 |
+
"eval_steps_per_second": 8.583,
|
| 389 |
+
"step": 2000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 18.26086956521739,
|
| 393 |
+
"grad_norm": 4.607916355133057,
|
| 394 |
+
"learning_rate": 5.2777777777777785e-06,
|
| 395 |
+
"loss": 0.0673,
|
| 396 |
+
"step": 2100
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 18.26086956521739,
|
| 400 |
+
"eval_accuracy": 0.9320467518347377,
|
| 401 |
+
"eval_f1_score": 0.925184613434992,
|
| 402 |
+
"eval_loss": 0.2904324531555176,
|
| 403 |
+
"eval_precision": 0.9332741752610002,
|
| 404 |
+
"eval_recall": 0.9184445089519725,
|
| 405 |
+
"eval_runtime": 6.7294,
|
| 406 |
+
"eval_samples_per_second": 546.702,
|
| 407 |
+
"eval_steps_per_second": 8.619,
|
| 408 |
+
"step": 2100
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 19.130434782608695,
|
| 412 |
+
"grad_norm": 6.327251434326172,
|
| 413 |
+
"learning_rate": 5e-06,
|
| 414 |
+
"loss": 0.0571,
|
| 415 |
+
"step": 2200
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 19.130434782608695,
|
| 419 |
+
"eval_accuracy": 0.9293286219081273,
|
| 420 |
+
"eval_f1_score": 0.9221668516434853,
|
| 421 |
+
"eval_loss": 0.3166205883026123,
|
| 422 |
+
"eval_precision": 0.925137476734381,
|
| 423 |
+
"eval_recall": 0.920952609526737,
|
| 424 |
+
"eval_runtime": 6.7037,
|
| 425 |
+
"eval_samples_per_second": 548.799,
|
| 426 |
+
"eval_steps_per_second": 8.652,
|
| 427 |
+
"step": 2200
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 20.0,
|
| 431 |
+
"grad_norm": 9.082805633544922,
|
| 432 |
+
"learning_rate": 4.722222222222222e-06,
|
| 433 |
+
"loss": 0.0561,
|
| 434 |
+
"step": 2300
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 20.0,
|
| 438 |
+
"eval_accuracy": 0.9317749388420766,
|
| 439 |
+
"eval_f1_score": 0.9221280725480369,
|
| 440 |
+
"eval_loss": 0.2922111749649048,
|
| 441 |
+
"eval_precision": 0.9150274852978553,
|
| 442 |
+
"eval_recall": 0.9297539237688469,
|
| 443 |
+
"eval_runtime": 6.7503,
|
| 444 |
+
"eval_samples_per_second": 545.017,
|
| 445 |
+
"eval_steps_per_second": 8.592,
|
| 446 |
+
"step": 2300
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 20.869565217391305,
|
| 450 |
+
"grad_norm": 5.283856391906738,
|
| 451 |
+
"learning_rate": 4.444444444444444e-06,
|
| 452 |
+
"loss": 0.0511,
|
| 453 |
+
"step": 2400
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 20.869565217391305,
|
| 457 |
+
"eval_accuracy": 0.9315031258494156,
|
| 458 |
+
"eval_f1_score": 0.9190655007648246,
|
| 459 |
+
"eval_loss": 0.29927295446395874,
|
| 460 |
+
"eval_precision": 0.9088064828335735,
|
| 461 |
+
"eval_recall": 0.9303236730969998,
|
| 462 |
+
"eval_runtime": 6.7281,
|
| 463 |
+
"eval_samples_per_second": 546.809,
|
| 464 |
+
"eval_steps_per_second": 8.621,
|
| 465 |
+
"step": 2400
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 21.73913043478261,
|
| 469 |
+
"grad_norm": 6.0074896812438965,
|
| 470 |
+
"learning_rate": 4.166666666666667e-06,
|
| 471 |
+
"loss": 0.0442,
|
| 472 |
+
"step": 2500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 21.73913043478261,
|
| 476 |
+
"eval_accuracy": 0.9266104919815167,
|
| 477 |
+
"eval_f1_score": 0.9161795338292905,
|
| 478 |
+
"eval_loss": 0.32011494040489197,
|
| 479 |
+
"eval_precision": 0.9060451440252857,
|
| 480 |
+
"eval_recall": 0.9280493422296427,
|
| 481 |
+
"eval_runtime": 6.7127,
|
| 482 |
+
"eval_samples_per_second": 548.067,
|
| 483 |
+
"eval_steps_per_second": 8.64,
|
| 484 |
+
"step": 2500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 22.608695652173914,
|
| 488 |
+
"grad_norm": 3.1078407764434814,
|
| 489 |
+
"learning_rate": 3.88888888888889e-06,
|
| 490 |
+
"loss": 0.0447,
|
| 491 |
+
"step": 2600
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 22.608695652173914,
|
| 495 |
+
"eval_accuracy": 0.928241369937483,
|
| 496 |
+
"eval_f1_score": 0.9137497551284842,
|
| 497 |
+
"eval_loss": 0.3155056834220886,
|
| 498 |
+
"eval_precision": 0.9009580466238951,
|
| 499 |
+
"eval_recall": 0.9281730038314259,
|
| 500 |
+
"eval_runtime": 6.7337,
|
| 501 |
+
"eval_samples_per_second": 546.354,
|
| 502 |
+
"eval_steps_per_second": 8.613,
|
| 503 |
+
"step": 2600
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 23.47826086956522,
|
| 507 |
+
"grad_norm": 2.9584195613861084,
|
| 508 |
+
"learning_rate": 3.6111111111111115e-06,
|
| 509 |
+
"loss": 0.0415,
|
| 510 |
+
"step": 2700
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 23.47826086956522,
|
| 514 |
+
"eval_accuracy": 0.9334058167980429,
|
| 515 |
+
"eval_f1_score": 0.9226018260362496,
|
| 516 |
+
"eval_loss": 0.30177852511405945,
|
| 517 |
+
"eval_precision": 0.9185179495480513,
|
| 518 |
+
"eval_recall": 0.9269833265460256,
|
| 519 |
+
"eval_runtime": 6.7411,
|
| 520 |
+
"eval_samples_per_second": 545.757,
|
| 521 |
+
"eval_steps_per_second": 8.604,
|
| 522 |
+
"step": 2700
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 24.347826086956523,
|
| 526 |
+
"grad_norm": 12.190321922302246,
|
| 527 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 528 |
+
"loss": 0.0359,
|
| 529 |
+
"step": 2800
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 24.347826086956523,
|
| 533 |
+
"eval_accuracy": 0.9298722478934494,
|
| 534 |
+
"eval_f1_score": 0.9177278989948806,
|
| 535 |
+
"eval_loss": 0.31918126344680786,
|
| 536 |
+
"eval_precision": 0.9062664068560837,
|
| 537 |
+
"eval_recall": 0.9308234663752396,
|
| 538 |
+
"eval_runtime": 6.7802,
|
| 539 |
+
"eval_samples_per_second": 542.606,
|
| 540 |
+
"eval_steps_per_second": 8.554,
|
| 541 |
+
"step": 2800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 25.217391304347824,
|
| 545 |
+
"grad_norm": 0.2598835527896881,
|
| 546 |
+
"learning_rate": 3.055555555555556e-06,
|
| 547 |
+
"loss": 0.0369,
|
| 548 |
+
"step": 2900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 25.217391304347824,
|
| 552 |
+
"eval_accuracy": 0.933677629790704,
|
| 553 |
+
"eval_f1_score": 0.9210521238209074,
|
| 554 |
+
"eval_loss": 0.3063570559024811,
|
| 555 |
+
"eval_precision": 0.9140578271273506,
|
| 556 |
+
"eval_recall": 0.9285610502121662,
|
| 557 |
+
"eval_runtime": 6.7729,
|
| 558 |
+
"eval_samples_per_second": 543.197,
|
| 559 |
+
"eval_steps_per_second": 8.564,
|
| 560 |
+
"step": 2900
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 26.08695652173913,
|
| 564 |
+
"grad_norm": 0.24433408677577972,
|
| 565 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 566 |
+
"loss": 0.0296,
|
| 567 |
+
"step": 3000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 26.08695652173913,
|
| 571 |
+
"eval_accuracy": 0.9328621908127208,
|
| 572 |
+
"eval_f1_score": 0.9237047805131925,
|
| 573 |
+
"eval_loss": 0.311038613319397,
|
| 574 |
+
"eval_precision": 0.9198460229141495,
|
| 575 |
+
"eval_recall": 0.9279424126946928,
|
| 576 |
+
"eval_runtime": 6.8161,
|
| 577 |
+
"eval_samples_per_second": 539.754,
|
| 578 |
+
"eval_steps_per_second": 8.509,
|
| 579 |
+
"step": 3000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 26.08695652173913,
|
| 583 |
+
"step": 3000,
|
| 584 |
+
"total_flos": 6579999363349350.0,
|
| 585 |
+
"train_loss": 0.31640464369455973,
|
| 586 |
+
"train_runtime": 3238.6183,
|
| 587 |
+
"train_samples_per_second": 158.092,
|
| 588 |
+
"train_steps_per_second": 1.235
|
| 589 |
+
}
|
| 590 |
+
],
|
| 591 |
+
"logging_steps": 100,
|
| 592 |
+
"max_steps": 4000,
|
| 593 |
+
"num_input_tokens_seen": 0,
|
| 594 |
+
"num_train_epochs": 35,
|
| 595 |
+
"save_steps": 100,
|
| 596 |
+
"total_flos": 6579999363349350.0,
|
| 597 |
+
"train_batch_size": 64,
|
| 598 |
+
"trial_name": null,
|
| 599 |
+
"trial_params": null
|
| 600 |
+
}
|