Commit
•
c31436d
1
Parent(s):
accf0e5
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 +562 -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: 0.3274
|
| 22 |
+
- Accuracy: 0.9268
|
| 23 |
+
- F1 Score: 0.8919
|
| 24 |
+
- Recall: 0.8944
|
| 25 |
+
- Precision: 0.8897
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.79,
|
| 3 |
+
"eval_accuracy": 0.9267926792679267,
|
| 4 |
+
"eval_f1_score": 0.8919460068914353,
|
| 5 |
+
"eval_loss": 0.3273985981941223,
|
| 6 |
+
"eval_precision": 0.8896937401913322,
|
| 7 |
+
"eval_recall": 0.8944103789209331,
|
| 8 |
+
"eval_runtime": 6.5937,
|
| 9 |
+
"eval_samples_per_second": 505.479,
|
| 10 |
+
"eval_steps_per_second": 8.038,
|
| 11 |
+
"total_flos": 6555293809710912.0,
|
| 12 |
+
"train_loss": 0.28978255237851824,
|
| 13 |
+
"train_runtime": 2707.5005,
|
| 14 |
+
"train_samples_per_second": 189.104,
|
| 15 |
+
"train_steps_per_second": 1.477
|
| 16 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.79,
|
| 3 |
+
"eval_accuracy": 0.9267926792679267,
|
| 4 |
+
"eval_f1_score": 0.8919460068914353,
|
| 5 |
+
"eval_loss": 0.3273985981941223,
|
| 6 |
+
"eval_precision": 0.8896937401913322,
|
| 7 |
+
"eval_recall": 0.8944103789209331,
|
| 8 |
+
"eval_runtime": 6.5937,
|
| 9 |
+
"eval_samples_per_second": 505.479,
|
| 10 |
+
"eval_steps_per_second": 8.038
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.79,
|
| 3 |
+
"total_flos": 6555293809710912.0,
|
| 4 |
+
"train_loss": 0.28978255237851824,
|
| 5 |
+
"train_runtime": 2707.5005,
|
| 6 |
+
"train_samples_per_second": 189.104,
|
| 7 |
+
"train_steps_per_second": 1.477
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,562 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8919460068914353,
|
| 3 |
+
"best_model_checkpoint": "cls_comment-phobert-base-v2-v2.4.1/checkpoint-2300",
|
| 4 |
+
"epoch": 26.794258373205743,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 2800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.96,
|
| 13 |
+
"grad_norm": 1.6857236623764038,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 1.7049,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.96,
|
| 20 |
+
"eval_accuracy": 0.46144614461446143,
|
| 21 |
+
"eval_f1_score": 0.10801682151937281,
|
| 22 |
+
"eval_loss": 1.4949835538864136,
|
| 23 |
+
"eval_precision": 0.21007122078443174,
|
| 24 |
+
"eval_recall": 0.16814650388457272,
|
| 25 |
+
"eval_runtime": 6.4303,
|
| 26 |
+
"eval_samples_per_second": 518.327,
|
| 27 |
+
"eval_steps_per_second": 8.242,
|
| 28 |
+
"step": 100
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.91,
|
| 32 |
+
"grad_norm": 2.8904168605804443,
|
| 33 |
+
"learning_rate": 5e-06,
|
| 34 |
+
"loss": 1.3066,
|
| 35 |
+
"step": 200
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.91,
|
| 39 |
+
"eval_accuracy": 0.6597659765976598,
|
| 40 |
+
"eval_f1_score": 0.24932737263473323,
|
| 41 |
+
"eval_loss": 1.0450588464736938,
|
| 42 |
+
"eval_precision": 0.21503708017040588,
|
| 43 |
+
"eval_recall": 0.29703415957578,
|
| 44 |
+
"eval_runtime": 6.2935,
|
| 45 |
+
"eval_samples_per_second": 529.594,
|
| 46 |
+
"eval_steps_per_second": 8.421,
|
| 47 |
+
"step": 200
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.87,
|
| 51 |
+
"grad_norm": 4.716286659240723,
|
| 52 |
+
"learning_rate": 7.500000000000001e-06,
|
| 53 |
+
"loss": 0.9457,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.87,
|
| 58 |
+
"eval_accuracy": 0.7971797179717972,
|
| 59 |
+
"eval_f1_score": 0.5218513395163904,
|
| 60 |
+
"eval_loss": 0.7491273283958435,
|
| 61 |
+
"eval_precision": 0.5237982031957714,
|
| 62 |
+
"eval_recall": 0.5230368404187595,
|
| 63 |
+
"eval_runtime": 6.3214,
|
| 64 |
+
"eval_samples_per_second": 527.257,
|
| 65 |
+
"eval_steps_per_second": 8.384,
|
| 66 |
+
"step": 300
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 3.83,
|
| 70 |
+
"grad_norm": 4.255033016204834,
|
| 71 |
+
"learning_rate": 1e-05,
|
| 72 |
+
"loss": 0.6975,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 3.83,
|
| 77 |
+
"eval_accuracy": 0.8496849684968497,
|
| 78 |
+
"eval_f1_score": 0.569984189542809,
|
| 79 |
+
"eval_loss": 0.5573880672454834,
|
| 80 |
+
"eval_precision": 0.7142864553492309,
|
| 81 |
+
"eval_recall": 0.5934564957773306,
|
| 82 |
+
"eval_runtime": 6.2462,
|
| 83 |
+
"eval_samples_per_second": 533.607,
|
| 84 |
+
"eval_steps_per_second": 8.485,
|
| 85 |
+
"step": 400
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4.78,
|
| 89 |
+
"grad_norm": 9.066823959350586,
|
| 90 |
+
"learning_rate": 9.722222222222223e-06,
|
| 91 |
+
"loss": 0.5187,
|
| 92 |
+
"step": 500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.78,
|
| 96 |
+
"eval_accuracy": 0.8664866486648665,
|
| 97 |
+
"eval_f1_score": 0.6685229055473728,
|
| 98 |
+
"eval_loss": 0.46813032031059265,
|
| 99 |
+
"eval_precision": 0.7076541248010603,
|
| 100 |
+
"eval_recall": 0.6591572058707156,
|
| 101 |
+
"eval_runtime": 6.2485,
|
| 102 |
+
"eval_samples_per_second": 533.408,
|
| 103 |
+
"eval_steps_per_second": 8.482,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 5.74,
|
| 108 |
+
"grad_norm": 5.799560070037842,
|
| 109 |
+
"learning_rate": 9.444444444444445e-06,
|
| 110 |
+
"loss": 0.4183,
|
| 111 |
+
"step": 600
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 5.74,
|
| 115 |
+
"eval_accuracy": 0.8820882088208821,
|
| 116 |
+
"eval_f1_score": 0.774744299508991,
|
| 117 |
+
"eval_loss": 0.4121437668800354,
|
| 118 |
+
"eval_precision": 0.8760939894193666,
|
| 119 |
+
"eval_recall": 0.7478297020405164,
|
| 120 |
+
"eval_runtime": 6.289,
|
| 121 |
+
"eval_samples_per_second": 529.974,
|
| 122 |
+
"eval_steps_per_second": 8.427,
|
| 123 |
+
"step": 600
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 6.7,
|
| 127 |
+
"grad_norm": 5.299034595489502,
|
| 128 |
+
"learning_rate": 9.166666666666666e-06,
|
| 129 |
+
"loss": 0.3323,
|
| 130 |
+
"step": 700
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 6.7,
|
| 134 |
+
"eval_accuracy": 0.903990399039904,
|
| 135 |
+
"eval_f1_score": 0.850497109426447,
|
| 136 |
+
"eval_loss": 0.34878164529800415,
|
| 137 |
+
"eval_precision": 0.8647331643827426,
|
| 138 |
+
"eval_recall": 0.839118025290519,
|
| 139 |
+
"eval_runtime": 6.1906,
|
| 140 |
+
"eval_samples_per_second": 538.4,
|
| 141 |
+
"eval_steps_per_second": 8.561,
|
| 142 |
+
"step": 700
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 7.66,
|
| 146 |
+
"grad_norm": 5.325865268707275,
|
| 147 |
+
"learning_rate": 8.888888888888888e-06,
|
| 148 |
+
"loss": 0.2705,
|
| 149 |
+
"step": 800
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 7.66,
|
| 153 |
+
"eval_accuracy": 0.9123912391239124,
|
| 154 |
+
"eval_f1_score": 0.8679529123789608,
|
| 155 |
+
"eval_loss": 0.3178523778915405,
|
| 156 |
+
"eval_precision": 0.8683198068711735,
|
| 157 |
+
"eval_recall": 0.8693736512924785,
|
| 158 |
+
"eval_runtime": 6.2028,
|
| 159 |
+
"eval_samples_per_second": 537.341,
|
| 160 |
+
"eval_steps_per_second": 8.545,
|
| 161 |
+
"step": 800
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 8.61,
|
| 165 |
+
"grad_norm": 4.130426406860352,
|
| 166 |
+
"learning_rate": 8.611111111111112e-06,
|
| 167 |
+
"loss": 0.229,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 8.61,
|
| 172 |
+
"eval_accuracy": 0.915991599159916,
|
| 173 |
+
"eval_f1_score": 0.8739115802564847,
|
| 174 |
+
"eval_loss": 0.3108988404273987,
|
| 175 |
+
"eval_precision": 0.8704138816567074,
|
| 176 |
+
"eval_recall": 0.8778451486716982,
|
| 177 |
+
"eval_runtime": 6.2053,
|
| 178 |
+
"eval_samples_per_second": 537.123,
|
| 179 |
+
"eval_steps_per_second": 8.541,
|
| 180 |
+
"step": 900
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 9.57,
|
| 184 |
+
"grad_norm": 4.939031600952148,
|
| 185 |
+
"learning_rate": 8.333333333333334e-06,
|
| 186 |
+
"loss": 0.1964,
|
| 187 |
+
"step": 1000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 9.57,
|
| 191 |
+
"eval_accuracy": 0.9174917491749175,
|
| 192 |
+
"eval_f1_score": 0.8775661449676492,
|
| 193 |
+
"eval_loss": 0.3028014600276947,
|
| 194 |
+
"eval_precision": 0.8740691720600018,
|
| 195 |
+
"eval_recall": 0.8812896711088255,
|
| 196 |
+
"eval_runtime": 6.2156,
|
| 197 |
+
"eval_samples_per_second": 536.232,
|
| 198 |
+
"eval_steps_per_second": 8.527,
|
| 199 |
+
"step": 1000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 10.53,
|
| 203 |
+
"grad_norm": 7.8568878173828125,
|
| 204 |
+
"learning_rate": 8.055555555555557e-06,
|
| 205 |
+
"loss": 0.1771,
|
| 206 |
+
"step": 1100
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 10.53,
|
| 210 |
+
"eval_accuracy": 0.918091809180918,
|
| 211 |
+
"eval_f1_score": 0.8807318795078937,
|
| 212 |
+
"eval_loss": 0.30322423577308655,
|
| 213 |
+
"eval_precision": 0.8743317066593773,
|
| 214 |
+
"eval_recall": 0.8877139802470034,
|
| 215 |
+
"eval_runtime": 6.2526,
|
| 216 |
+
"eval_samples_per_second": 533.058,
|
| 217 |
+
"eval_steps_per_second": 8.476,
|
| 218 |
+
"step": 1100
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 11.48,
|
| 222 |
+
"grad_norm": 8.988609313964844,
|
| 223 |
+
"learning_rate": 7.77777777777778e-06,
|
| 224 |
+
"loss": 0.1518,
|
| 225 |
+
"step": 1200
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 11.48,
|
| 229 |
+
"eval_accuracy": 0.9165916591659166,
|
| 230 |
+
"eval_f1_score": 0.8762082418137522,
|
| 231 |
+
"eval_loss": 0.3150691092014313,
|
| 232 |
+
"eval_precision": 0.8828474311875153,
|
| 233 |
+
"eval_recall": 0.8702178521387077,
|
| 234 |
+
"eval_runtime": 6.2783,
|
| 235 |
+
"eval_samples_per_second": 530.878,
|
| 236 |
+
"eval_steps_per_second": 8.442,
|
| 237 |
+
"step": 1200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 12.44,
|
| 241 |
+
"grad_norm": 6.535445690155029,
|
| 242 |
+
"learning_rate": 7.500000000000001e-06,
|
| 243 |
+
"loss": 0.1368,
|
| 244 |
+
"step": 1300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 12.44,
|
| 248 |
+
"eval_accuracy": 0.9213921392139214,
|
| 249 |
+
"eval_f1_score": 0.8794257945528638,
|
| 250 |
+
"eval_loss": 0.29376497864723206,
|
| 251 |
+
"eval_precision": 0.8789472587468637,
|
| 252 |
+
"eval_recall": 0.8800344348044677,
|
| 253 |
+
"eval_runtime": 6.2765,
|
| 254 |
+
"eval_samples_per_second": 531.025,
|
| 255 |
+
"eval_steps_per_second": 8.444,
|
| 256 |
+
"step": 1300
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 13.4,
|
| 260 |
+
"grad_norm": 6.200416564941406,
|
| 261 |
+
"learning_rate": 7.222222222222223e-06,
|
| 262 |
+
"loss": 0.1116,
|
| 263 |
+
"step": 1400
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 13.4,
|
| 267 |
+
"eval_accuracy": 0.9204920492049204,
|
| 268 |
+
"eval_f1_score": 0.8794630237866472,
|
| 269 |
+
"eval_loss": 0.29709574580192566,
|
| 270 |
+
"eval_precision": 0.8776282905981169,
|
| 271 |
+
"eval_recall": 0.881459841742562,
|
| 272 |
+
"eval_runtime": 6.1898,
|
| 273 |
+
"eval_samples_per_second": 538.463,
|
| 274 |
+
"eval_steps_per_second": 8.562,
|
| 275 |
+
"step": 1400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 14.35,
|
| 279 |
+
"grad_norm": 5.921875,
|
| 280 |
+
"learning_rate": 6.944444444444445e-06,
|
| 281 |
+
"loss": 0.1136,
|
| 282 |
+
"step": 1500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 14.35,
|
| 286 |
+
"eval_accuracy": 0.9234923492349235,
|
| 287 |
+
"eval_f1_score": 0.8857825928642494,
|
| 288 |
+
"eval_loss": 0.30112606287002563,
|
| 289 |
+
"eval_precision": 0.8894251615389424,
|
| 290 |
+
"eval_recall": 0.8825136450024287,
|
| 291 |
+
"eval_runtime": 6.2125,
|
| 292 |
+
"eval_samples_per_second": 536.5,
|
| 293 |
+
"eval_steps_per_second": 8.531,
|
| 294 |
+
"step": 1500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 15.31,
|
| 298 |
+
"grad_norm": 13.020410537719727,
|
| 299 |
+
"learning_rate": 6.666666666666667e-06,
|
| 300 |
+
"loss": 0.094,
|
| 301 |
+
"step": 1600
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 15.31,
|
| 305 |
+
"eval_accuracy": 0.9267926792679267,
|
| 306 |
+
"eval_f1_score": 0.8891204106050078,
|
| 307 |
+
"eval_loss": 0.2937377691268921,
|
| 308 |
+
"eval_precision": 0.8855321931614877,
|
| 309 |
+
"eval_recall": 0.8933184630720276,
|
| 310 |
+
"eval_runtime": 6.2824,
|
| 311 |
+
"eval_samples_per_second": 530.532,
|
| 312 |
+
"eval_steps_per_second": 8.436,
|
| 313 |
+
"step": 1600
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 16.27,
|
| 317 |
+
"grad_norm": 6.572815418243408,
|
| 318 |
+
"learning_rate": 6.3888888888888885e-06,
|
| 319 |
+
"loss": 0.0905,
|
| 320 |
+
"step": 1700
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 16.27,
|
| 324 |
+
"eval_accuracy": 0.9264926492649265,
|
| 325 |
+
"eval_f1_score": 0.8849818091608542,
|
| 326 |
+
"eval_loss": 0.30491721630096436,
|
| 327 |
+
"eval_precision": 0.8885868021143595,
|
| 328 |
+
"eval_recall": 0.8819482912765212,
|
| 329 |
+
"eval_runtime": 6.1862,
|
| 330 |
+
"eval_samples_per_second": 538.781,
|
| 331 |
+
"eval_steps_per_second": 8.567,
|
| 332 |
+
"step": 1700
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 17.22,
|
| 336 |
+
"grad_norm": 11.92620849609375,
|
| 337 |
+
"learning_rate": 6.111111111111112e-06,
|
| 338 |
+
"loss": 0.0838,
|
| 339 |
+
"step": 1800
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 17.22,
|
| 343 |
+
"eval_accuracy": 0.9243924392439244,
|
| 344 |
+
"eval_f1_score": 0.8823178893108493,
|
| 345 |
+
"eval_loss": 0.30605384707450867,
|
| 346 |
+
"eval_precision": 0.8783548672468386,
|
| 347 |
+
"eval_recall": 0.8869437658317699,
|
| 348 |
+
"eval_runtime": 6.213,
|
| 349 |
+
"eval_samples_per_second": 536.453,
|
| 350 |
+
"eval_steps_per_second": 8.53,
|
| 351 |
+
"step": 1800
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 18.18,
|
| 355 |
+
"grad_norm": 6.095840930938721,
|
| 356 |
+
"learning_rate": 5.833333333333334e-06,
|
| 357 |
+
"loss": 0.0749,
|
| 358 |
+
"step": 1900
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 18.18,
|
| 362 |
+
"eval_accuracy": 0.9204920492049204,
|
| 363 |
+
"eval_f1_score": 0.8771352609895953,
|
| 364 |
+
"eval_loss": 0.3274901509284973,
|
| 365 |
+
"eval_precision": 0.8717396476070715,
|
| 366 |
+
"eval_recall": 0.883927539576653,
|
| 367 |
+
"eval_runtime": 6.221,
|
| 368 |
+
"eval_samples_per_second": 535.769,
|
| 369 |
+
"eval_steps_per_second": 8.52,
|
| 370 |
+
"step": 1900
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 19.14,
|
| 374 |
+
"grad_norm": 6.83023738861084,
|
| 375 |
+
"learning_rate": 5.555555555555557e-06,
|
| 376 |
+
"loss": 0.0686,
|
| 377 |
+
"step": 2000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 19.14,
|
| 381 |
+
"eval_accuracy": 0.9294929492949295,
|
| 382 |
+
"eval_f1_score": 0.8915069438815207,
|
| 383 |
+
"eval_loss": 0.3092004358768463,
|
| 384 |
+
"eval_precision": 0.8845623861216612,
|
| 385 |
+
"eval_recall": 0.8989552312613528,
|
| 386 |
+
"eval_runtime": 6.2372,
|
| 387 |
+
"eval_samples_per_second": 534.375,
|
| 388 |
+
"eval_steps_per_second": 8.497,
|
| 389 |
+
"step": 2000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 20.1,
|
| 393 |
+
"grad_norm": 4.03262186050415,
|
| 394 |
+
"learning_rate": 5.2777777777777785e-06,
|
| 395 |
+
"loss": 0.0669,
|
| 396 |
+
"step": 2100
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 20.1,
|
| 400 |
+
"eval_accuracy": 0.924992499249925,
|
| 401 |
+
"eval_f1_score": 0.8835986677013299,
|
| 402 |
+
"eval_loss": 0.3168080151081085,
|
| 403 |
+
"eval_precision": 0.8824571821499484,
|
| 404 |
+
"eval_recall": 0.8848697486050657,
|
| 405 |
+
"eval_runtime": 6.2215,
|
| 406 |
+
"eval_samples_per_second": 535.72,
|
| 407 |
+
"eval_steps_per_second": 8.519,
|
| 408 |
+
"step": 2100
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 21.05,
|
| 412 |
+
"grad_norm": 1.4791430234909058,
|
| 413 |
+
"learning_rate": 5e-06,
|
| 414 |
+
"loss": 0.0582,
|
| 415 |
+
"step": 2200
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 21.05,
|
| 419 |
+
"eval_accuracy": 0.9234923492349235,
|
| 420 |
+
"eval_f1_score": 0.8762756882804612,
|
| 421 |
+
"eval_loss": 0.333869606256485,
|
| 422 |
+
"eval_precision": 0.8630877156327875,
|
| 423 |
+
"eval_recall": 0.8925977028354047,
|
| 424 |
+
"eval_runtime": 6.2137,
|
| 425 |
+
"eval_samples_per_second": 536.399,
|
| 426 |
+
"eval_steps_per_second": 8.53,
|
| 427 |
+
"step": 2200
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 22.01,
|
| 431 |
+
"grad_norm": 1.3378651142120361,
|
| 432 |
+
"learning_rate": 4.722222222222222e-06,
|
| 433 |
+
"loss": 0.0516,
|
| 434 |
+
"step": 2300
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 22.01,
|
| 438 |
+
"eval_accuracy": 0.9267926792679267,
|
| 439 |
+
"eval_f1_score": 0.8919460068914353,
|
| 440 |
+
"eval_loss": 0.3273985981941223,
|
| 441 |
+
"eval_precision": 0.8896937401913322,
|
| 442 |
+
"eval_recall": 0.8944103789209331,
|
| 443 |
+
"eval_runtime": 6.2539,
|
| 444 |
+
"eval_samples_per_second": 532.947,
|
| 445 |
+
"eval_steps_per_second": 8.475,
|
| 446 |
+
"step": 2300
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 22.97,
|
| 450 |
+
"grad_norm": 2.3011834621429443,
|
| 451 |
+
"learning_rate": 4.444444444444444e-06,
|
| 452 |
+
"loss": 0.0543,
|
| 453 |
+
"step": 2400
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 22.97,
|
| 457 |
+
"eval_accuracy": 0.9294929492949295,
|
| 458 |
+
"eval_f1_score": 0.8912698081733191,
|
| 459 |
+
"eval_loss": 0.32296448945999146,
|
| 460 |
+
"eval_precision": 0.8946222329092709,
|
| 461 |
+
"eval_recall": 0.8882296547805383,
|
| 462 |
+
"eval_runtime": 6.2569,
|
| 463 |
+
"eval_samples_per_second": 532.688,
|
| 464 |
+
"eval_steps_per_second": 8.471,
|
| 465 |
+
"step": 2400
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 23.92,
|
| 469 |
+
"grad_norm": 0.30257853865623474,
|
| 470 |
+
"learning_rate": 4.166666666666667e-06,
|
| 471 |
+
"loss": 0.0435,
|
| 472 |
+
"step": 2500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 23.92,
|
| 476 |
+
"eval_accuracy": 0.9252925292529253,
|
| 477 |
+
"eval_f1_score": 0.8806373103770997,
|
| 478 |
+
"eval_loss": 0.3364492952823639,
|
| 479 |
+
"eval_precision": 0.8917652898214161,
|
| 480 |
+
"eval_recall": 0.8705348522415428,
|
| 481 |
+
"eval_runtime": 6.2476,
|
| 482 |
+
"eval_samples_per_second": 533.488,
|
| 483 |
+
"eval_steps_per_second": 8.483,
|
| 484 |
+
"step": 2500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 24.88,
|
| 488 |
+
"grad_norm": 4.352519512176514,
|
| 489 |
+
"learning_rate": 3.88888888888889e-06,
|
| 490 |
+
"loss": 0.0405,
|
| 491 |
+
"step": 2600
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 24.88,
|
| 495 |
+
"eval_accuracy": 0.9240924092409241,
|
| 496 |
+
"eval_f1_score": 0.8816233559879275,
|
| 497 |
+
"eval_loss": 0.34915250539779663,
|
| 498 |
+
"eval_precision": 0.8818560471877731,
|
| 499 |
+
"eval_recall": 0.8820867223989518,
|
| 500 |
+
"eval_runtime": 6.2343,
|
| 501 |
+
"eval_samples_per_second": 534.62,
|
| 502 |
+
"eval_steps_per_second": 8.501,
|
| 503 |
+
"step": 2600
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 25.84,
|
| 507 |
+
"grad_norm": 1.6296316385269165,
|
| 508 |
+
"learning_rate": 3.6111111111111115e-06,
|
| 509 |
+
"loss": 0.0398,
|
| 510 |
+
"step": 2700
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 25.84,
|
| 514 |
+
"eval_accuracy": 0.9237923792379238,
|
| 515 |
+
"eval_f1_score": 0.8799317994806395,
|
| 516 |
+
"eval_loss": 0.3558264374732971,
|
| 517 |
+
"eval_precision": 0.8807466782442214,
|
| 518 |
+
"eval_recall": 0.879644875034168,
|
| 519 |
+
"eval_runtime": 6.2137,
|
| 520 |
+
"eval_samples_per_second": 536.396,
|
| 521 |
+
"eval_steps_per_second": 8.53,
|
| 522 |
+
"step": 2700
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 26.79,
|
| 526 |
+
"grad_norm": 0.5989682078361511,
|
| 527 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 528 |
+
"loss": 0.0363,
|
| 529 |
+
"step": 2800
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 26.79,
|
| 533 |
+
"eval_accuracy": 0.9222922292229223,
|
| 534 |
+
"eval_f1_score": 0.8741787015976087,
|
| 535 |
+
"eval_loss": 0.36046338081359863,
|
| 536 |
+
"eval_precision": 0.8698137895606535,
|
| 537 |
+
"eval_recall": 0.8795311139961707,
|
| 538 |
+
"eval_runtime": 6.2828,
|
| 539 |
+
"eval_samples_per_second": 530.494,
|
| 540 |
+
"eval_steps_per_second": 8.436,
|
| 541 |
+
"step": 2800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 26.79,
|
| 545 |
+
"step": 2800,
|
| 546 |
+
"total_flos": 6555293809710912.0,
|
| 547 |
+
"train_loss": 0.28978255237851824,
|
| 548 |
+
"train_runtime": 2707.5005,
|
| 549 |
+
"train_samples_per_second": 189.104,
|
| 550 |
+
"train_steps_per_second": 1.477
|
| 551 |
+
}
|
| 552 |
+
],
|
| 553 |
+
"logging_steps": 100,
|
| 554 |
+
"max_steps": 4000,
|
| 555 |
+
"num_input_tokens_seen": 0,
|
| 556 |
+
"num_train_epochs": 39,
|
| 557 |
+
"save_steps": 100,
|
| 558 |
+
"total_flos": 6555293809710912.0,
|
| 559 |
+
"train_batch_size": 64,
|
| 560 |
+
"trial_name": null,
|
| 561 |
+
"trial_params": null
|
| 562 |
+
}
|