Commit
•
c721c7b
1
Parent(s):
7174fd4
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 +410 -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: 0.6177
|
| 22 |
+
- Accuracy: 0.9241
|
| 23 |
+
- F1 Score: 0.8919
|
| 24 |
+
- Recall: 0.8930
|
| 25 |
+
- Precision: 0.8911
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.14,
|
| 3 |
+
"eval_accuracy": 0.9240924092409241,
|
| 4 |
+
"eval_f1_score": 0.8919400441736002,
|
| 5 |
+
"eval_loss": 0.6176608204841614,
|
| 6 |
+
"eval_precision": 0.8911083019322857,
|
| 7 |
+
"eval_recall": 0.8930303467825517,
|
| 8 |
+
"eval_runtime": 6.2449,
|
| 9 |
+
"eval_samples_per_second": 533.714,
|
| 10 |
+
"eval_steps_per_second": 8.487,
|
| 11 |
+
"total_flos": 4678479916291584.0,
|
| 12 |
+
"train_loss": 0.6728471851348877,
|
| 13 |
+
"train_runtime": 1885.5017,
|
| 14 |
+
"train_samples_per_second": 271.546,
|
| 15 |
+
"train_steps_per_second": 2.121
|
| 16 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.14,
|
| 3 |
+
"eval_accuracy": 0.9240924092409241,
|
| 4 |
+
"eval_f1_score": 0.8919400441736002,
|
| 5 |
+
"eval_loss": 0.6176608204841614,
|
| 6 |
+
"eval_precision": 0.8911083019322857,
|
| 7 |
+
"eval_recall": 0.8930303467825517,
|
| 8 |
+
"eval_runtime": 6.2449,
|
| 9 |
+
"eval_samples_per_second": 533.714,
|
| 10 |
+
"eval_steps_per_second": 8.487
|
| 11 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.14,
|
| 3 |
+
"total_flos": 4678479916291584.0,
|
| 4 |
+
"train_loss": 0.6728471851348877,
|
| 5 |
+
"train_runtime": 1885.5017,
|
| 6 |
+
"train_samples_per_second": 271.546,
|
| 7 |
+
"train_steps_per_second": 2.121
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,410 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8919400441736002,
|
| 3 |
+
"best_model_checkpoint": "cls_comment-phobert-base-v2-v2.4/checkpoint-1500",
|
| 4 |
+
"epoch": 19.138755980861244,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 2000,
|
| 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": 1.4111480712890625,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 1.7008,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.96,
|
| 20 |
+
"eval_accuracy": 0.46234623462346236,
|
| 21 |
+
"eval_f1_score": 0.10914902653284753,
|
| 22 |
+
"eval_loss": 1.5259455442428589,
|
| 23 |
+
"eval_precision": 0.22956187895212285,
|
| 24 |
+
"eval_recall": 0.16870144284128744,
|
| 25 |
+
"eval_runtime": 5.8062,
|
| 26 |
+
"eval_samples_per_second": 574.041,
|
| 27 |
+
"eval_steps_per_second": 9.128,
|
| 28 |
+
"step": 100
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.91,
|
| 32 |
+
"grad_norm": 3.172757148742676,
|
| 33 |
+
"learning_rate": 5e-06,
|
| 34 |
+
"loss": 1.4089,
|
| 35 |
+
"step": 200
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.91,
|
| 39 |
+
"eval_accuracy": 0.6567656765676567,
|
| 40 |
+
"eval_f1_score": 0.24987557712552566,
|
| 41 |
+
"eval_loss": 1.1875011920928955,
|
| 42 |
+
"eval_precision": 0.216976056542887,
|
| 43 |
+
"eval_recall": 0.2948041271015744,
|
| 44 |
+
"eval_runtime": 6.0549,
|
| 45 |
+
"eval_samples_per_second": 550.46,
|
| 46 |
+
"eval_steps_per_second": 8.753,
|
| 47 |
+
"step": 200
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.87,
|
| 51 |
+
"grad_norm": 4.23789644241333,
|
| 52 |
+
"learning_rate": 7.500000000000001e-06,
|
| 53 |
+
"loss": 1.0776,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.87,
|
| 58 |
+
"eval_accuracy": 0.8037803780378038,
|
| 59 |
+
"eval_f1_score": 0.5304195101025111,
|
| 60 |
+
"eval_loss": 0.9009036421775818,
|
| 61 |
+
"eval_precision": 0.5333440517242575,
|
| 62 |
+
"eval_recall": 0.5308669916547205,
|
| 63 |
+
"eval_runtime": 6.0211,
|
| 64 |
+
"eval_samples_per_second": 553.553,
|
| 65 |
+
"eval_steps_per_second": 8.802,
|
| 66 |
+
"step": 300
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 3.83,
|
| 70 |
+
"grad_norm": 2.9036450386047363,
|
| 71 |
+
"learning_rate": 1e-05,
|
| 72 |
+
"loss": 0.8625,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 3.83,
|
| 77 |
+
"eval_accuracy": 0.8574857485748575,
|
| 78 |
+
"eval_f1_score": 0.6321452494594356,
|
| 79 |
+
"eval_loss": 0.7617138624191284,
|
| 80 |
+
"eval_precision": 0.7107045568063487,
|
| 81 |
+
"eval_recall": 0.6372368264036259,
|
| 82 |
+
"eval_runtime": 6.0254,
|
| 83 |
+
"eval_samples_per_second": 553.154,
|
| 84 |
+
"eval_steps_per_second": 8.796,
|
| 85 |
+
"step": 400
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4.78,
|
| 89 |
+
"grad_norm": 4.648315906524658,
|
| 90 |
+
"learning_rate": 9.722222222222223e-06,
|
| 91 |
+
"loss": 0.7245,
|
| 92 |
+
"step": 500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.78,
|
| 96 |
+
"eval_accuracy": 0.8817881788178817,
|
| 97 |
+
"eval_f1_score": 0.7282115691268598,
|
| 98 |
+
"eval_loss": 0.689365804195404,
|
| 99 |
+
"eval_precision": 0.8802547295767839,
|
| 100 |
+
"eval_recall": 0.7213609117839314,
|
| 101 |
+
"eval_runtime": 6.0362,
|
| 102 |
+
"eval_samples_per_second": 552.166,
|
| 103 |
+
"eval_steps_per_second": 8.78,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 5.74,
|
| 108 |
+
"grad_norm": 5.482232093811035,
|
| 109 |
+
"learning_rate": 9.444444444444445e-06,
|
| 110 |
+
"loss": 0.6573,
|
| 111 |
+
"step": 600
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 5.74,
|
| 115 |
+
"eval_accuracy": 0.8967896789678967,
|
| 116 |
+
"eval_f1_score": 0.8406229534596467,
|
| 117 |
+
"eval_loss": 0.6651212573051453,
|
| 118 |
+
"eval_precision": 0.8770203982052687,
|
| 119 |
+
"eval_recall": 0.8212623271516084,
|
| 120 |
+
"eval_runtime": 6.089,
|
| 121 |
+
"eval_samples_per_second": 547.376,
|
| 122 |
+
"eval_steps_per_second": 8.704,
|
| 123 |
+
"step": 600
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 6.7,
|
| 127 |
+
"grad_norm": 4.107127666473389,
|
| 128 |
+
"learning_rate": 9.166666666666666e-06,
|
| 129 |
+
"loss": 0.6082,
|
| 130 |
+
"step": 700
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 6.7,
|
| 134 |
+
"eval_accuracy": 0.9078907890789079,
|
| 135 |
+
"eval_f1_score": 0.8629840378859187,
|
| 136 |
+
"eval_loss": 0.6335443258285522,
|
| 137 |
+
"eval_precision": 0.8594850910270425,
|
| 138 |
+
"eval_recall": 0.866740275628739,
|
| 139 |
+
"eval_runtime": 6.012,
|
| 140 |
+
"eval_samples_per_second": 554.389,
|
| 141 |
+
"eval_steps_per_second": 8.816,
|
| 142 |
+
"step": 700
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 7.66,
|
| 146 |
+
"grad_norm": 4.067190647125244,
|
| 147 |
+
"learning_rate": 8.888888888888888e-06,
|
| 148 |
+
"loss": 0.5674,
|
| 149 |
+
"step": 800
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 7.66,
|
| 153 |
+
"eval_accuracy": 0.9105910591059105,
|
| 154 |
+
"eval_f1_score": 0.8692050710637084,
|
| 155 |
+
"eval_loss": 0.636346697807312,
|
| 156 |
+
"eval_precision": 0.8621114850690574,
|
| 157 |
+
"eval_recall": 0.8795216165515357,
|
| 158 |
+
"eval_runtime": 6.0275,
|
| 159 |
+
"eval_samples_per_second": 552.965,
|
| 160 |
+
"eval_steps_per_second": 8.793,
|
| 161 |
+
"step": 800
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 8.61,
|
| 165 |
+
"grad_norm": 3.8194425106048584,
|
| 166 |
+
"learning_rate": 8.611111111111112e-06,
|
| 167 |
+
"loss": 0.5477,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 8.61,
|
| 172 |
+
"eval_accuracy": 0.9150915091509151,
|
| 173 |
+
"eval_f1_score": 0.8776204392865118,
|
| 174 |
+
"eval_loss": 0.6268996596336365,
|
| 175 |
+
"eval_precision": 0.8876632968569648,
|
| 176 |
+
"eval_recall": 0.8693279668307071,
|
| 177 |
+
"eval_runtime": 5.982,
|
| 178 |
+
"eval_samples_per_second": 557.174,
|
| 179 |
+
"eval_steps_per_second": 8.86,
|
| 180 |
+
"step": 900
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 9.57,
|
| 184 |
+
"grad_norm": 4.406583309173584,
|
| 185 |
+
"learning_rate": 8.333333333333334e-06,
|
| 186 |
+
"loss": 0.5256,
|
| 187 |
+
"step": 1000
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 9.57,
|
| 191 |
+
"eval_accuracy": 0.9204920492049204,
|
| 192 |
+
"eval_f1_score": 0.8834661339438045,
|
| 193 |
+
"eval_loss": 0.6178370714187622,
|
| 194 |
+
"eval_precision": 0.8826304228202296,
|
| 195 |
+
"eval_recall": 0.8848781731721993,
|
| 196 |
+
"eval_runtime": 6.0125,
|
| 197 |
+
"eval_samples_per_second": 554.349,
|
| 198 |
+
"eval_steps_per_second": 8.815,
|
| 199 |
+
"step": 1000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 10.53,
|
| 203 |
+
"grad_norm": 3.69624400138855,
|
| 204 |
+
"learning_rate": 8.055555555555557e-06,
|
| 205 |
+
"loss": 0.5148,
|
| 206 |
+
"step": 1100
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 10.53,
|
| 210 |
+
"eval_accuracy": 0.9198919891989199,
|
| 211 |
+
"eval_f1_score": 0.879559155366957,
|
| 212 |
+
"eval_loss": 0.6214434504508972,
|
| 213 |
+
"eval_precision": 0.8761821574730276,
|
| 214 |
+
"eval_recall": 0.8838583582269935,
|
| 215 |
+
"eval_runtime": 6.0361,
|
| 216 |
+
"eval_samples_per_second": 552.174,
|
| 217 |
+
"eval_steps_per_second": 8.78,
|
| 218 |
+
"step": 1100
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 11.48,
|
| 222 |
+
"grad_norm": 4.3708343505859375,
|
| 223 |
+
"learning_rate": 7.77777777777778e-06,
|
| 224 |
+
"loss": 0.4999,
|
| 225 |
+
"step": 1200
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 11.48,
|
| 229 |
+
"eval_accuracy": 0.9228922892289229,
|
| 230 |
+
"eval_f1_score": 0.8856181382103513,
|
| 231 |
+
"eval_loss": 0.6158125400543213,
|
| 232 |
+
"eval_precision": 0.8862286567723959,
|
| 233 |
+
"eval_recall": 0.8853146392819923,
|
| 234 |
+
"eval_runtime": 5.9971,
|
| 235 |
+
"eval_samples_per_second": 555.768,
|
| 236 |
+
"eval_steps_per_second": 8.838,
|
| 237 |
+
"step": 1200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 12.44,
|
| 241 |
+
"grad_norm": 5.073043346405029,
|
| 242 |
+
"learning_rate": 7.500000000000001e-06,
|
| 243 |
+
"loss": 0.4916,
|
| 244 |
+
"step": 1300
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 12.44,
|
| 248 |
+
"eval_accuracy": 0.9231923192319232,
|
| 249 |
+
"eval_f1_score": 0.8839178584078228,
|
| 250 |
+
"eval_loss": 0.6185785531997681,
|
| 251 |
+
"eval_precision": 0.8887592838927109,
|
| 252 |
+
"eval_recall": 0.8794527947878894,
|
| 253 |
+
"eval_runtime": 6.0868,
|
| 254 |
+
"eval_samples_per_second": 547.575,
|
| 255 |
+
"eval_steps_per_second": 8.707,
|
| 256 |
+
"step": 1300
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 13.4,
|
| 260 |
+
"grad_norm": 3.3584494590759277,
|
| 261 |
+
"learning_rate": 7.222222222222223e-06,
|
| 262 |
+
"loss": 0.479,
|
| 263 |
+
"step": 1400
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 13.4,
|
| 267 |
+
"eval_accuracy": 0.9201920192019202,
|
| 268 |
+
"eval_f1_score": 0.8847465693125924,
|
| 269 |
+
"eval_loss": 0.6285073161125183,
|
| 270 |
+
"eval_precision": 0.8863862882902076,
|
| 271 |
+
"eval_recall": 0.8833493957517122,
|
| 272 |
+
"eval_runtime": 6.0466,
|
| 273 |
+
"eval_samples_per_second": 551.218,
|
| 274 |
+
"eval_steps_per_second": 8.765,
|
| 275 |
+
"step": 1400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 14.35,
|
| 279 |
+
"grad_norm": 4.605256080627441,
|
| 280 |
+
"learning_rate": 6.944444444444445e-06,
|
| 281 |
+
"loss": 0.4812,
|
| 282 |
+
"step": 1500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 14.35,
|
| 286 |
+
"eval_accuracy": 0.9240924092409241,
|
| 287 |
+
"eval_f1_score": 0.8919400441736002,
|
| 288 |
+
"eval_loss": 0.6176608204841614,
|
| 289 |
+
"eval_precision": 0.8911083019322857,
|
| 290 |
+
"eval_recall": 0.8930303467825517,
|
| 291 |
+
"eval_runtime": 6.0144,
|
| 292 |
+
"eval_samples_per_second": 554.172,
|
| 293 |
+
"eval_steps_per_second": 8.812,
|
| 294 |
+
"step": 1500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 15.31,
|
| 298 |
+
"grad_norm": 5.669193744659424,
|
| 299 |
+
"learning_rate": 6.666666666666667e-06,
|
| 300 |
+
"loss": 0.4667,
|
| 301 |
+
"step": 1600
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 15.31,
|
| 305 |
+
"eval_accuracy": 0.9255925592559255,
|
| 306 |
+
"eval_f1_score": 0.8847926962066609,
|
| 307 |
+
"eval_loss": 0.6205594539642334,
|
| 308 |
+
"eval_precision": 0.8843481979823253,
|
| 309 |
+
"eval_recall": 0.8852982243831954,
|
| 310 |
+
"eval_runtime": 6.1133,
|
| 311 |
+
"eval_samples_per_second": 545.203,
|
| 312 |
+
"eval_steps_per_second": 8.67,
|
| 313 |
+
"step": 1600
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 16.27,
|
| 317 |
+
"grad_norm": 3.3217363357543945,
|
| 318 |
+
"learning_rate": 6.3888888888888885e-06,
|
| 319 |
+
"loss": 0.4668,
|
| 320 |
+
"step": 1700
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 16.27,
|
| 324 |
+
"eval_accuracy": 0.9264926492649265,
|
| 325 |
+
"eval_f1_score": 0.8854482169609333,
|
| 326 |
+
"eval_loss": 0.6200627684593201,
|
| 327 |
+
"eval_precision": 0.8837445827045557,
|
| 328 |
+
"eval_recall": 0.8875863206208373,
|
| 329 |
+
"eval_runtime": 6.0633,
|
| 330 |
+
"eval_samples_per_second": 549.698,
|
| 331 |
+
"eval_steps_per_second": 8.741,
|
| 332 |
+
"step": 1700
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 17.22,
|
| 336 |
+
"grad_norm": 3.535698652267456,
|
| 337 |
+
"learning_rate": 6.111111111111112e-06,
|
| 338 |
+
"loss": 0.4635,
|
| 339 |
+
"step": 1800
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 17.22,
|
| 343 |
+
"eval_accuracy": 0.9252925292529253,
|
| 344 |
+
"eval_f1_score": 0.8901467035784716,
|
| 345 |
+
"eval_loss": 0.6252107620239258,
|
| 346 |
+
"eval_precision": 0.8926952965988483,
|
| 347 |
+
"eval_recall": 0.8877172530901484,
|
| 348 |
+
"eval_runtime": 6.0706,
|
| 349 |
+
"eval_samples_per_second": 549.043,
|
| 350 |
+
"eval_steps_per_second": 8.731,
|
| 351 |
+
"step": 1800
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 18.18,
|
| 355 |
+
"grad_norm": 3.0793726444244385,
|
| 356 |
+
"learning_rate": 5.833333333333334e-06,
|
| 357 |
+
"loss": 0.4593,
|
| 358 |
+
"step": 1900
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 18.18,
|
| 362 |
+
"eval_accuracy": 0.9273927392739274,
|
| 363 |
+
"eval_f1_score": 0.88911901951679,
|
| 364 |
+
"eval_loss": 0.626419723033905,
|
| 365 |
+
"eval_precision": 0.8887473645082232,
|
| 366 |
+
"eval_recall": 0.8899116044774821,
|
| 367 |
+
"eval_runtime": 6.0336,
|
| 368 |
+
"eval_samples_per_second": 552.41,
|
| 369 |
+
"eval_steps_per_second": 8.784,
|
| 370 |
+
"step": 1900
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 19.14,
|
| 374 |
+
"grad_norm": 0.5877706408500671,
|
| 375 |
+
"learning_rate": 5.555555555555557e-06,
|
| 376 |
+
"loss": 0.4538,
|
| 377 |
+
"step": 2000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 19.14,
|
| 381 |
+
"eval_accuracy": 0.9264926492649265,
|
| 382 |
+
"eval_f1_score": 0.8890948519955958,
|
| 383 |
+
"eval_loss": 0.6227503418922424,
|
| 384 |
+
"eval_precision": 0.8870090750160463,
|
| 385 |
+
"eval_recall": 0.8913327030956307,
|
| 386 |
+
"eval_runtime": 6.0553,
|
| 387 |
+
"eval_samples_per_second": 550.427,
|
| 388 |
+
"eval_steps_per_second": 8.753,
|
| 389 |
+
"step": 2000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 19.14,
|
| 393 |
+
"step": 2000,
|
| 394 |
+
"total_flos": 4678479916291584.0,
|
| 395 |
+
"train_loss": 0.6728471851348877,
|
| 396 |
+
"train_runtime": 1885.5017,
|
| 397 |
+
"train_samples_per_second": 271.546,
|
| 398 |
+
"train_steps_per_second": 2.121
|
| 399 |
+
}
|
| 400 |
+
],
|
| 401 |
+
"logging_steps": 100,
|
| 402 |
+
"max_steps": 4000,
|
| 403 |
+
"num_input_tokens_seen": 0,
|
| 404 |
+
"num_train_epochs": 39,
|
| 405 |
+
"save_steps": 100,
|
| 406 |
+
"total_flos": 4678479916291584.0,
|
| 407 |
+
"train_batch_size": 64,
|
| 408 |
+
"trial_name": null,
|
| 409 |
+
"trial_params": null
|
| 410 |
+
}
|