End of training
Browse files- README.md +9 -9
- all_results.json +12 -12
- runs/Jun15_00-58-48_92b2e0e6fb20/events.out.tfevents.1749949423.92b2e0e6fb20.2194.14 +3 -0
- test_results.json +13 -13
- trainer_state.json +692 -563
README.md
CHANGED
|
@@ -16,15 +16,15 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [maximuspowers/bert-philosophy-adapted](https://huggingface.co/maximuspowers/bert-philosophy-adapted) on the None dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
-
- Exact Match Accuracy: 0.
|
| 21 |
-
- Macro Precision: 0.
|
| 22 |
-
- Macro Recall: 0.
|
| 23 |
-
- Macro F1: 0.
|
| 24 |
-
- Micro Precision: 0.
|
| 25 |
-
- Micro Recall: 0.
|
| 26 |
-
- Micro F1: 0.
|
| 27 |
-
- Hamming Loss: 0.
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [maximuspowers/bert-philosophy-adapted](https://huggingface.co/maximuspowers/bert-philosophy-adapted) on the None dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.4468
|
| 20 |
+
- Exact Match Accuracy: 0.425
|
| 21 |
+
- Macro Precision: 0.3078
|
| 22 |
+
- Macro Recall: 0.2221
|
| 23 |
+
- Macro F1: 0.2248
|
| 24 |
+
- Micro Precision: 0.8966
|
| 25 |
+
- Micro Recall: 0.4561
|
| 26 |
+
- Micro F1: 0.6047
|
| 27 |
+
- Hamming Loss: 0.05
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
all_results.json
CHANGED
|
@@ -1,17 +1,17 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
-
"eval_exact_match_accuracy": 0.
|
| 4 |
-
"eval_hamming_loss": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_macro_f1": 0.
|
| 7 |
-
"eval_macro_precision": 0.
|
| 8 |
-
"eval_macro_recall": 0.
|
| 9 |
-
"eval_micro_f1": 0.
|
| 10 |
-
"eval_micro_precision": 0.
|
| 11 |
-
"eval_micro_recall": 0.
|
| 12 |
-
"eval_runtime": 0.
|
| 13 |
-
"eval_samples_per_second":
|
| 14 |
-
"eval_steps_per_second": 23.
|
| 15 |
"total_flos": 0.0,
|
| 16 |
"train_loss": 0.8574352493286133,
|
| 17 |
"train_runtime": 257.7927,
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
+
"eval_exact_match_accuracy": 0.425,
|
| 4 |
+
"eval_hamming_loss": 0.05,
|
| 5 |
+
"eval_loss": 0.44675666093826294,
|
| 6 |
+
"eval_macro_f1": 0.22477092910529442,
|
| 7 |
+
"eval_macro_precision": 0.307843137254902,
|
| 8 |
+
"eval_macro_recall": 0.22205882352941175,
|
| 9 |
+
"eval_micro_f1": 0.6046511627906976,
|
| 10 |
+
"eval_micro_precision": 0.896551724137931,
|
| 11 |
+
"eval_micro_recall": 0.45614035087719296,
|
| 12 |
+
"eval_runtime": 0.2085,
|
| 13 |
+
"eval_samples_per_second": 191.847,
|
| 14 |
+
"eval_steps_per_second": 23.981,
|
| 15 |
"total_flos": 0.0,
|
| 16 |
"train_loss": 0.8574352493286133,
|
| 17 |
"train_runtime": 257.7927,
|
runs/Jun15_00-58-48_92b2e0e6fb20/events.out.tfevents.1749949423.92b2e0e6fb20.2194.14
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5681d8937701cc06c9ca22fe68ebe13b29d8ca175164cd965fa953de1ae2047
|
| 3 |
+
size 5853
|
test_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_exact_match_accuracy": 0.
|
| 4 |
-
"eval_hamming_loss": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_macro_f1": 0.
|
| 7 |
-
"eval_macro_precision": 0.
|
| 8 |
-
"eval_macro_recall": 0.
|
| 9 |
-
"eval_micro_f1": 0.
|
| 10 |
-
"eval_micro_precision": 0.
|
| 11 |
-
"eval_micro_recall": 0.
|
| 12 |
-
"eval_runtime": 0.
|
| 13 |
-
"eval_samples_per_second":
|
| 14 |
-
"eval_steps_per_second": 23.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_exact_match_accuracy": 0.425,
|
| 4 |
+
"eval_hamming_loss": 0.05,
|
| 5 |
+
"eval_loss": 0.44675666093826294,
|
| 6 |
+
"eval_macro_f1": 0.22477092910529442,
|
| 7 |
+
"eval_macro_precision": 0.307843137254902,
|
| 8 |
+
"eval_macro_recall": 0.22205882352941175,
|
| 9 |
+
"eval_micro_f1": 0.6046511627906976,
|
| 10 |
+
"eval_micro_precision": 0.896551724137931,
|
| 11 |
+
"eval_micro_recall": 0.45614035087719296,
|
| 12 |
+
"eval_runtime": 0.2085,
|
| 13 |
+
"eval_samples_per_second": 191.847,
|
| 14 |
+
"eval_steps_per_second": 23.981
|
| 15 |
}
|
trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
"best_metric": 0.42105263157894735,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -12,1179 +12,1308 @@
|
|
| 12 |
{
|
| 13 |
"epoch": 0,
|
| 14 |
"step": 0,
|
| 15 |
-
"train/classification_loss": 0.
|
| 16 |
-
"train/contrastive_loss": 9.
|
| 17 |
-
"train/negative_loss": 9.
|
| 18 |
"train/num_negatives": 46,
|
| 19 |
"train/num_positives": 10,
|
| 20 |
-
"train/positive_loss":
|
| 21 |
-
"train/total_loss": 2.
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0,
|
| 25 |
"step": 0,
|
| 26 |
-
"train/classification_loss": 0.
|
| 27 |
-
"train/contrastive_loss":
|
| 28 |
-
"train/negative_loss":
|
| 29 |
"train/num_negatives": 46,
|
| 30 |
"train/num_positives": 10,
|
| 31 |
-
"train/positive_loss":
|
| 32 |
-
"train/total_loss": 2.
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 2.5,
|
| 36 |
-
"grad_norm": 18.
|
| 37 |
-
"learning_rate": 9.
|
| 38 |
-
"loss": 3.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.5,
|
| 43 |
"step": 50,
|
| 44 |
-
"train/classification_loss": 0.
|
| 45 |
-
"train/contrastive_loss": 1.
|
| 46 |
-
"train/negative_loss": 1.
|
| 47 |
"train/num_negatives": 38,
|
| 48 |
"train/num_positives": 18,
|
| 49 |
-
"train/positive_loss": 0.
|
| 50 |
-
"train/total_loss": 1.
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 2.5,
|
| 54 |
"step": 50,
|
| 55 |
-
"train/classification_loss": 0.
|
| 56 |
-
"train/contrastive_loss": 1.
|
| 57 |
-
"train/negative_loss": 1.
|
| 58 |
"train/num_negatives": 50,
|
| 59 |
"train/num_positives": 6,
|
| 60 |
-
"train/positive_loss": 0.
|
| 61 |
-
"train/total_loss":
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 5.0,
|
| 65 |
-
"grad_norm":
|
| 66 |
-
"learning_rate": 1.
|
| 67 |
-
"loss": 1.
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 5.0,
|
| 72 |
"step": 100,
|
| 73 |
-
"train/classification_loss": 0.
|
| 74 |
-
"train/contrastive_loss": 2.
|
| 75 |
-
"train/negative_loss": 1.
|
| 76 |
"train/num_negatives": 44,
|
| 77 |
"train/num_positives": 12,
|
| 78 |
-
"train/positive_loss": 0.
|
| 79 |
-
"train/total_loss": 0.
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 5.0,
|
| 83 |
"step": 100,
|
| 84 |
-
"train/classification_loss": 0.
|
| 85 |
-
"train/contrastive_loss":
|
| 86 |
-
"train/negative_loss":
|
| 87 |
"train/num_negatives": 44,
|
| 88 |
"train/num_positives": 12,
|
| 89 |
-
"train/positive_loss": 0.
|
| 90 |
-
"train/total_loss": 1.
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 5.0,
|
| 94 |
"step": 100,
|
| 95 |
-
"train/classification_loss": 0.
|
| 96 |
-
"train/contrastive_loss": 1.
|
| 97 |
-
"train/negative_loss": 1.
|
| 98 |
"train/num_negatives": 52,
|
| 99 |
"train/num_positives": 4,
|
| 100 |
-
"train/positive_loss": 0.
|
| 101 |
-
"train/total_loss": 0.
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 5.0,
|
| 105 |
"step": 100,
|
| 106 |
-
"train/classification_loss": 0.
|
| 107 |
-
"train/contrastive_loss": 3.
|
| 108 |
-
"train/negative_loss": 2.
|
| 109 |
"train/num_negatives": 48,
|
| 110 |
"train/num_positives": 8,
|
| 111 |
-
"train/positive_loss": 0.
|
| 112 |
-
"train/total_loss": 1.
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 5.0,
|
| 116 |
"step": 100,
|
| 117 |
-
"train/classification_loss": 0.
|
| 118 |
-
"train/contrastive_loss":
|
| 119 |
-
"train/negative_loss": 1.
|
| 120 |
"train/num_negatives": 46,
|
| 121 |
"train/num_positives": 8,
|
| 122 |
-
"train/positive_loss":
|
| 123 |
-
"train/total_loss": 0.
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 5.0,
|
| 127 |
"eval_exact_match_accuracy": 0.0,
|
| 128 |
-
"eval_hamming_loss": 0.
|
| 129 |
-
"eval_loss":
|
| 130 |
-
"eval_macro_f1": 0.
|
| 131 |
-
"eval_macro_precision": 0.
|
| 132 |
-
"eval_macro_recall": 0.
|
| 133 |
-
"eval_micro_f1": 0.
|
| 134 |
-
"eval_micro_precision":
|
| 135 |
-
"eval_micro_recall": 0.
|
| 136 |
-
"eval_runtime": 0.
|
| 137 |
-
"eval_samples_per_second":
|
| 138 |
-
"eval_steps_per_second":
|
| 139 |
"step": 100
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 5.0,
|
| 143 |
"step": 100,
|
| 144 |
-
"train/classification_loss": 0.
|
| 145 |
-
"train/contrastive_loss": 1.
|
| 146 |
-
"train/negative_loss":
|
| 147 |
"train/num_negatives": 42,
|
| 148 |
"train/num_positives": 12,
|
| 149 |
-
"train/positive_loss": 0.
|
| 150 |
-
"train/total_loss": 0.
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 5.0,
|
| 154 |
"step": 100,
|
| 155 |
-
"train/classification_loss": 0.
|
| 156 |
-
"train/contrastive_loss": 0.
|
| 157 |
-
"train/negative_loss": 0.
|
| 158 |
"train/num_negatives": 40,
|
| 159 |
"train/num_positives": 16,
|
| 160 |
-
"train/positive_loss": 0.
|
| 161 |
-
"train/total_loss": 0.
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 7.5,
|
| 165 |
-
"grad_norm":
|
| 166 |
-
"learning_rate": 1.
|
| 167 |
-
"loss": 1.
|
| 168 |
"step": 150
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 7.5,
|
| 172 |
"step": 150,
|
| 173 |
-
"train/classification_loss": 0.
|
| 174 |
-
"train/contrastive_loss": 1.
|
| 175 |
-
"train/negative_loss": 0.
|
| 176 |
"train/num_negatives": 44,
|
| 177 |
"train/num_positives": 8,
|
| 178 |
-
"train/positive_loss": 0.
|
| 179 |
-
"train/total_loss": 0.
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 7.5,
|
| 183 |
"step": 150,
|
| 184 |
-
"train/classification_loss": 0.
|
| 185 |
-
"train/contrastive_loss":
|
| 186 |
-
"train/negative_loss":
|
| 187 |
"train/num_negatives": 44,
|
| 188 |
"train/num_positives": 10,
|
| 189 |
-
"train/positive_loss": 0.
|
| 190 |
-
"train/total_loss": 0.
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 10.0,
|
| 194 |
-
"grad_norm":
|
| 195 |
-
"learning_rate": 1.
|
| 196 |
-
"loss": 1.
|
| 197 |
"step": 200
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 10.0,
|
| 201 |
"step": 200,
|
| 202 |
-
"train/classification_loss": 0.
|
| 203 |
-
"train/contrastive_loss":
|
| 204 |
-
"train/negative_loss":
|
| 205 |
"train/num_negatives": 44,
|
| 206 |
"train/num_positives": 12,
|
| 207 |
-
"train/positive_loss": 0.
|
| 208 |
-
"train/total_loss": 0.
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"epoch": 10.0,
|
| 212 |
"step": 200,
|
| 213 |
-
"train/classification_loss": 0.
|
| 214 |
-
"train/contrastive_loss": 3.
|
| 215 |
-
"train/negative_loss": 3.
|
| 216 |
"train/num_negatives": 44,
|
| 217 |
"train/num_positives": 12,
|
| 218 |
-
"train/positive_loss": 0.
|
| 219 |
-
"train/total_loss":
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 10.0,
|
| 223 |
"step": 200,
|
| 224 |
-
"train/classification_loss": 0.
|
| 225 |
-
"train/contrastive_loss": 2.
|
| 226 |
-
"train/negative_loss": 2.
|
| 227 |
"train/num_negatives": 52,
|
| 228 |
"train/num_positives": 4,
|
| 229 |
-
"train/positive_loss": 0.
|
| 230 |
-
"train/total_loss": 0.
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 10.0,
|
| 234 |
"step": 200,
|
| 235 |
-
"train/classification_loss": 0.
|
| 236 |
-
"train/contrastive_loss": 3.
|
| 237 |
-
"train/negative_loss":
|
| 238 |
"train/num_negatives": 48,
|
| 239 |
"train/num_positives": 8,
|
| 240 |
-
"train/positive_loss":
|
| 241 |
-
"train/total_loss": 1.
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 10.0,
|
| 245 |
"step": 200,
|
| 246 |
-
"train/classification_loss": 0.
|
| 247 |
-
"train/contrastive_loss":
|
| 248 |
-
"train/negative_loss": 1.
|
| 249 |
"train/num_negatives": 46,
|
| 250 |
"train/num_positives": 8,
|
| 251 |
-
"train/positive_loss": 0.
|
| 252 |
-
"train/total_loss": 0.
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 10.0,
|
| 256 |
"eval_exact_match_accuracy": 0.0,
|
| 257 |
-
"eval_hamming_loss": 0.
|
| 258 |
-
"eval_loss": 0.
|
| 259 |
-
"eval_macro_f1": 0.
|
| 260 |
-
"eval_macro_precision": 0.
|
| 261 |
-
"eval_macro_recall": 0.
|
| 262 |
-
"eval_micro_f1": 0.
|
| 263 |
-
"eval_micro_precision":
|
| 264 |
-
"eval_micro_recall": 0.
|
| 265 |
-
"eval_runtime": 0.
|
| 266 |
-
"eval_samples_per_second":
|
| 267 |
-
"eval_steps_per_second":
|
| 268 |
"step": 200
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 10.0,
|
| 272 |
"step": 200,
|
| 273 |
-
"train/classification_loss": 0.
|
| 274 |
-
"train/contrastive_loss": 1.
|
| 275 |
-
"train/negative_loss": 0.
|
| 276 |
"train/num_negatives": 42,
|
| 277 |
"train/num_positives": 14,
|
| 278 |
-
"train/positive_loss": 0.
|
| 279 |
-
"train/total_loss": 0.
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 10.0,
|
| 283 |
"step": 200,
|
| 284 |
-
"train/classification_loss": 0.
|
| 285 |
-
"train/contrastive_loss":
|
| 286 |
-
"train/negative_loss":
|
| 287 |
"train/num_negatives": 42,
|
| 288 |
"train/num_positives": 14,
|
| 289 |
-
"train/positive_loss": 0.
|
| 290 |
-
"train/total_loss": 0.
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 12.5,
|
| 294 |
-
"grad_norm": 7.
|
| 295 |
-
"learning_rate": 1.
|
| 296 |
-
"loss": 1.
|
| 297 |
"step": 250
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 12.5,
|
| 301 |
"step": 250,
|
| 302 |
-
"train/classification_loss": 0.
|
| 303 |
-
"train/contrastive_loss":
|
| 304 |
-
"train/negative_loss":
|
| 305 |
"train/num_negatives": 44,
|
| 306 |
"train/num_positives": 12,
|
| 307 |
-
"train/positive_loss": 0.
|
| 308 |
-
"train/total_loss": 0.
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 12.5,
|
| 312 |
"step": 250,
|
| 313 |
-
"train/classification_loss": 0.
|
| 314 |
-
"train/contrastive_loss": 0.
|
| 315 |
-
"train/negative_loss": 0.
|
| 316 |
"train/num_negatives": 42,
|
| 317 |
"train/num_positives": 8,
|
| 318 |
-
"train/positive_loss": 0.
|
| 319 |
-
"train/total_loss": 0.
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 15.0,
|
| 323 |
-
"grad_norm":
|
| 324 |
-
"learning_rate": 1.
|
| 325 |
-
"loss": 0.
|
| 326 |
"step": 300
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 15.0,
|
| 330 |
"step": 300,
|
| 331 |
-
"train/classification_loss": 0.
|
| 332 |
-
"train/contrastive_loss":
|
| 333 |
-
"train/negative_loss": 2.
|
| 334 |
"train/num_negatives": 44,
|
| 335 |
"train/num_positives": 12,
|
| 336 |
-
"train/positive_loss": 0.
|
| 337 |
-
"train/total_loss": 0.
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 15.0,
|
| 341 |
"step": 300,
|
| 342 |
-
"train/classification_loss": 0.
|
| 343 |
-
"train/contrastive_loss":
|
| 344 |
-
"train/negative_loss":
|
| 345 |
"train/num_negatives": 44,
|
| 346 |
"train/num_positives": 12,
|
| 347 |
-
"train/positive_loss":
|
| 348 |
-
"train/total_loss": 1.
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 15.0,
|
| 352 |
"step": 300,
|
| 353 |
-
"train/classification_loss": 0.
|
| 354 |
-
"train/contrastive_loss":
|
| 355 |
-
"train/negative_loss": 2.
|
| 356 |
"train/num_negatives": 52,
|
| 357 |
"train/num_positives": 4,
|
| 358 |
-
"train/positive_loss":
|
| 359 |
-
"train/total_loss":
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 15.0,
|
| 363 |
"step": 300,
|
| 364 |
-
"train/classification_loss": 0.
|
| 365 |
-
"train/contrastive_loss":
|
| 366 |
-
"train/negative_loss": 4.
|
| 367 |
"train/num_negatives": 48,
|
| 368 |
"train/num_positives": 8,
|
| 369 |
-
"train/positive_loss": 0.
|
| 370 |
-
"train/total_loss": 1.
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 15.0,
|
| 374 |
"step": 300,
|
| 375 |
-
"train/classification_loss": 0.
|
| 376 |
-
"train/contrastive_loss":
|
| 377 |
-
"train/negative_loss": 1.
|
| 378 |
"train/num_negatives": 46,
|
| 379 |
"train/num_positives": 8,
|
| 380 |
-
"train/positive_loss": 0.
|
| 381 |
-
"train/total_loss": 0.
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 15.0,
|
| 385 |
"eval_exact_match_accuracy": 0.05,
|
| 386 |
-
"eval_hamming_loss": 0.
|
| 387 |
-
"eval_loss": 0.
|
| 388 |
-
"eval_macro_f1": 0.
|
| 389 |
-
"eval_macro_precision": 0.
|
| 390 |
"eval_macro_recall": 0.026737967914438502,
|
| 391 |
-
"eval_micro_f1": 0.
|
| 392 |
-
"eval_micro_precision":
|
| 393 |
"eval_micro_recall": 0.08771929824561403,
|
| 394 |
-
"eval_runtime": 0.
|
| 395 |
-
"eval_samples_per_second":
|
| 396 |
-
"eval_steps_per_second":
|
| 397 |
"step": 300
|
| 398 |
},
|
| 399 |
{
|
| 400 |
"epoch": 15.0,
|
| 401 |
"step": 300,
|
| 402 |
-
"train/classification_loss": 0.
|
| 403 |
-
"train/contrastive_loss": 0.
|
| 404 |
-
"train/negative_loss": 0.
|
| 405 |
"train/num_negatives": 32,
|
| 406 |
"train/num_positives": 20,
|
| 407 |
-
"train/positive_loss": 0.
|
| 408 |
-
"train/total_loss": 0.
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 15.0,
|
| 412 |
"step": 300,
|
| 413 |
-
"train/classification_loss": 0.
|
| 414 |
-
"train/contrastive_loss": 1.
|
| 415 |
-
"train/negative_loss": 1.
|
| 416 |
"train/num_negatives": 46,
|
| 417 |
"train/num_positives": 10,
|
| 418 |
-
"train/positive_loss": 0.
|
| 419 |
-
"train/total_loss": 0.
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 17.5,
|
| 423 |
-
"grad_norm":
|
| 424 |
-
"learning_rate": 1.
|
| 425 |
-
"loss": 0.
|
| 426 |
"step": 350
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 17.5,
|
| 430 |
"step": 350,
|
| 431 |
-
"train/classification_loss": 0.
|
| 432 |
-
"train/contrastive_loss": 0.
|
| 433 |
-
"train/negative_loss": 0.
|
| 434 |
"train/num_negatives": 42,
|
| 435 |
"train/num_positives": 14,
|
| 436 |
-
"train/positive_loss": 0.
|
| 437 |
-
"train/total_loss": 0.
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 17.5,
|
| 441 |
"step": 350,
|
| 442 |
-
"train/classification_loss": 0.
|
| 443 |
-
"train/contrastive_loss": 0.
|
| 444 |
-
"train/negative_loss": 0.
|
| 445 |
"train/num_negatives": 44,
|
| 446 |
"train/num_positives": 6,
|
| 447 |
-
"train/positive_loss": 0.
|
| 448 |
-
"train/total_loss": 0.
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 20.0,
|
| 452 |
-
"grad_norm":
|
| 453 |
-
"learning_rate": 1.
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 400
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 20.0,
|
| 459 |
"step": 400,
|
| 460 |
-
"train/classification_loss": 0.
|
| 461 |
-
"train/contrastive_loss":
|
| 462 |
-
"train/negative_loss": 1.
|
| 463 |
"train/num_negatives": 44,
|
| 464 |
"train/num_positives": 12,
|
| 465 |
-
"train/positive_loss": 0.
|
| 466 |
-
"train/total_loss": 0.
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 20.0,
|
| 470 |
"step": 400,
|
| 471 |
-
"train/classification_loss": 0.
|
| 472 |
-
"train/contrastive_loss": 3.
|
| 473 |
-
"train/negative_loss": 3.
|
| 474 |
"train/num_negatives": 44,
|
| 475 |
"train/num_positives": 12,
|
| 476 |
-
"train/positive_loss": 0.
|
| 477 |
-
"train/total_loss": 0.
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 20.0,
|
| 481 |
"step": 400,
|
| 482 |
-
"train/classification_loss": 0.
|
| 483 |
-
"train/contrastive_loss":
|
| 484 |
-
"train/negative_loss": 3.
|
| 485 |
"train/num_negatives": 52,
|
| 486 |
"train/num_positives": 4,
|
| 487 |
-
"train/positive_loss": 0.
|
| 488 |
-
"train/total_loss":
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 20.0,
|
| 492 |
"step": 400,
|
| 493 |
-
"train/classification_loss": 0.
|
| 494 |
-
"train/contrastive_loss":
|
| 495 |
-
"train/negative_loss":
|
| 496 |
"train/num_negatives": 48,
|
| 497 |
"train/num_positives": 8,
|
| 498 |
-
"train/positive_loss": 0.
|
| 499 |
-
"train/total_loss": 0.
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 20.0,
|
| 503 |
"step": 400,
|
| 504 |
-
"train/classification_loss": 0.
|
| 505 |
-
"train/contrastive_loss":
|
| 506 |
-
"train/negative_loss":
|
| 507 |
"train/num_negatives": 46,
|
| 508 |
"train/num_positives": 8,
|
| 509 |
-
"train/positive_loss": 0.
|
| 510 |
-
"train/total_loss": 0.
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 20.0,
|
| 514 |
"eval_exact_match_accuracy": 0.1,
|
| 515 |
"eval_hamming_loss": 0.075,
|
| 516 |
-
"eval_loss": 0.
|
| 517 |
-
"eval_macro_f1": 0.
|
| 518 |
-
"eval_macro_precision": 0.
|
| 519 |
-
"eval_macro_recall": 0.
|
| 520 |
-
"eval_micro_f1": 0.
|
| 521 |
-
"eval_micro_precision": 0.
|
| 522 |
-
"eval_micro_recall": 0.
|
| 523 |
-
"eval_runtime": 0.
|
| 524 |
-
"eval_samples_per_second":
|
| 525 |
-
"eval_steps_per_second": 24.
|
| 526 |
"step": 400
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 20.0,
|
| 530 |
"step": 400,
|
| 531 |
-
"train/classification_loss": 0.
|
| 532 |
-
"train/contrastive_loss": 0.
|
| 533 |
-
"train/negative_loss": 0.
|
| 534 |
"train/num_negatives": 42,
|
| 535 |
"train/num_positives": 10,
|
| 536 |
-
"train/positive_loss": 0.
|
| 537 |
-
"train/total_loss": 0.
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 20.0,
|
| 541 |
"step": 400,
|
| 542 |
-
"train/classification_loss": 0.
|
| 543 |
-
"train/contrastive_loss": 0.
|
| 544 |
-
"train/negative_loss": 0.
|
| 545 |
"train/num_negatives": 52,
|
| 546 |
"train/num_positives": 4,
|
| 547 |
-
"train/positive_loss": 0.
|
| 548 |
-
"train/total_loss": 0.
|
| 549 |
},
|
| 550 |
{
|
| 551 |
"epoch": 22.5,
|
| 552 |
-
"grad_norm":
|
| 553 |
-
"learning_rate": 1.
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 450
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 22.5,
|
| 559 |
"step": 450,
|
| 560 |
-
"train/classification_loss": 0.
|
| 561 |
-
"train/contrastive_loss":
|
| 562 |
-
"train/negative_loss":
|
| 563 |
"train/num_negatives": 50,
|
| 564 |
"train/num_positives": 4,
|
| 565 |
-
"train/positive_loss": 0.
|
| 566 |
-
"train/total_loss": 0.
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 22.5,
|
| 570 |
"step": 450,
|
| 571 |
-
"train/classification_loss": 0.
|
| 572 |
-
"train/contrastive_loss":
|
| 573 |
-
"train/negative_loss": 0.
|
| 574 |
"train/num_negatives": 50,
|
| 575 |
"train/num_positives": 6,
|
| 576 |
-
"train/positive_loss": 0.
|
| 577 |
-
"train/total_loss": 0.
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 25.0,
|
| 581 |
-
"grad_norm":
|
| 582 |
-
"learning_rate": 1.
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 25.0,
|
| 588 |
"step": 500,
|
| 589 |
-
"train/classification_loss": 0.
|
| 590 |
-
"train/contrastive_loss":
|
| 591 |
-
"train/negative_loss":
|
| 592 |
"train/num_negatives": 44,
|
| 593 |
"train/num_positives": 12,
|
| 594 |
-
"train/positive_loss": 0.
|
| 595 |
-
"train/total_loss": 0.
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 25.0,
|
| 599 |
"step": 500,
|
| 600 |
-
"train/classification_loss": 0.
|
| 601 |
-
"train/contrastive_loss": 2.
|
| 602 |
-
"train/negative_loss": 2.
|
| 603 |
"train/num_negatives": 44,
|
| 604 |
"train/num_positives": 12,
|
| 605 |
-
"train/positive_loss": 0.
|
| 606 |
-
"train/total_loss": 0.
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 25.0,
|
| 610 |
"step": 500,
|
| 611 |
-
"train/classification_loss": 0.
|
| 612 |
-
"train/contrastive_loss":
|
| 613 |
-
"train/negative_loss":
|
| 614 |
"train/num_negatives": 52,
|
| 615 |
"train/num_positives": 4,
|
| 616 |
-
"train/positive_loss": 1.
|
| 617 |
-
"train/total_loss": 1.
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 25.0,
|
| 621 |
"step": 500,
|
| 622 |
-
"train/classification_loss": 0.
|
| 623 |
-
"train/contrastive_loss":
|
| 624 |
-
"train/negative_loss":
|
| 625 |
"train/num_negatives": 48,
|
| 626 |
"train/num_positives": 8,
|
| 627 |
-
"train/positive_loss": 0.
|
| 628 |
-
"train/total_loss": 0.
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 25.0,
|
| 632 |
"step": 500,
|
| 633 |
-
"train/classification_loss": 0.
|
| 634 |
-
"train/contrastive_loss": 1.
|
| 635 |
-
"train/negative_loss": 0.
|
| 636 |
"train/num_negatives": 46,
|
| 637 |
"train/num_positives": 8,
|
| 638 |
-
"train/positive_loss":
|
| 639 |
-
"train/total_loss": 0.
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 25.0,
|
| 643 |
"eval_exact_match_accuracy": 0.175,
|
| 644 |
-
"eval_hamming_loss": 0.
|
| 645 |
-
"eval_loss": 0.
|
| 646 |
-
"eval_macro_f1": 0.
|
| 647 |
-
"eval_macro_precision": 0.
|
| 648 |
-
"eval_macro_recall": 0.
|
| 649 |
-
"eval_micro_f1": 0.
|
| 650 |
-
"eval_micro_precision": 0.
|
| 651 |
-
"eval_micro_recall": 0.
|
| 652 |
-
"eval_runtime": 0.
|
| 653 |
-
"eval_samples_per_second":
|
| 654 |
-
"eval_steps_per_second": 24.
|
| 655 |
"step": 500
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 25.0,
|
| 659 |
"step": 500,
|
| 660 |
-
"train/classification_loss": 0.
|
| 661 |
-
"train/contrastive_loss": 0.
|
| 662 |
-
"train/negative_loss": 0.
|
| 663 |
"train/num_negatives": 42,
|
| 664 |
"train/num_positives": 14,
|
| 665 |
-
"train/positive_loss": 0.
|
| 666 |
-
"train/total_loss": 0.
|
| 667 |
},
|
| 668 |
{
|
| 669 |
"epoch": 25.0,
|
| 670 |
"step": 500,
|
| 671 |
-
"train/classification_loss": 0.
|
| 672 |
-
"train/contrastive_loss":
|
| 673 |
-
"train/negative_loss": 0.
|
| 674 |
"train/num_negatives": 44,
|
| 675 |
"train/num_positives": 12,
|
| 676 |
-
"train/positive_loss": 0.
|
| 677 |
-
"train/total_loss": 0.
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 27.5,
|
| 681 |
-
"grad_norm": 6.
|
| 682 |
-
"learning_rate": 1.
|
| 683 |
-
"loss": 0.
|
| 684 |
"step": 550
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 27.5,
|
| 688 |
"step": 550,
|
| 689 |
-
"train/classification_loss": 0.
|
| 690 |
-
"train/contrastive_loss": 0.
|
| 691 |
-
"train/negative_loss": 0.
|
| 692 |
"train/num_negatives": 54,
|
| 693 |
"train/num_positives": 2,
|
| 694 |
-
"train/positive_loss":
|
| 695 |
-
"train/total_loss": 0.
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 27.5,
|
| 699 |
"step": 550,
|
| 700 |
-
"train/classification_loss": 0.
|
| 701 |
-
"train/contrastive_loss": 1.
|
| 702 |
-
"train/negative_loss": 1.
|
| 703 |
"train/num_negatives": 40,
|
| 704 |
"train/num_positives": 12,
|
| 705 |
-
"train/positive_loss": 0.
|
| 706 |
-
"train/total_loss": 0.
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 30.0,
|
| 710 |
-
"grad_norm":
|
| 711 |
-
"learning_rate":
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 600
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 30.0,
|
| 717 |
"step": 600,
|
| 718 |
-
"train/classification_loss": 0.
|
| 719 |
-
"train/contrastive_loss":
|
| 720 |
-
"train/negative_loss": 1.
|
| 721 |
"train/num_negatives": 44,
|
| 722 |
"train/num_positives": 12,
|
| 723 |
-
"train/positive_loss": 0.
|
| 724 |
-
"train/total_loss": 0.
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 30.0,
|
| 728 |
"step": 600,
|
| 729 |
-
"train/classification_loss": 0.
|
| 730 |
-
"train/contrastive_loss": 3.
|
| 731 |
-
"train/negative_loss": 3.
|
| 732 |
"train/num_negatives": 44,
|
| 733 |
"train/num_positives": 12,
|
| 734 |
-
"train/positive_loss": 0.
|
| 735 |
-
"train/total_loss": 0.
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 30.0,
|
| 739 |
"step": 600,
|
| 740 |
-
"train/classification_loss": 0.
|
| 741 |
-
"train/contrastive_loss":
|
| 742 |
-
"train/negative_loss":
|
| 743 |
"train/num_negatives": 52,
|
| 744 |
"train/num_positives": 4,
|
| 745 |
-
"train/positive_loss":
|
| 746 |
-
"train/total_loss":
|
| 747 |
},
|
| 748 |
{
|
| 749 |
"epoch": 30.0,
|
| 750 |
"step": 600,
|
| 751 |
-
"train/classification_loss": 0.
|
| 752 |
-
"train/contrastive_loss": 4.
|
| 753 |
-
"train/negative_loss":
|
| 754 |
"train/num_negatives": 48,
|
| 755 |
"train/num_positives": 8,
|
| 756 |
-
"train/positive_loss": 0.
|
| 757 |
-
"train/total_loss": 1.
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 30.0,
|
| 761 |
"step": 600,
|
| 762 |
-
"train/classification_loss": 0.
|
| 763 |
-
"train/contrastive_loss":
|
| 764 |
-
"train/negative_loss": 0.
|
| 765 |
"train/num_negatives": 46,
|
| 766 |
"train/num_positives": 8,
|
| 767 |
-
"train/positive_loss": 0.
|
| 768 |
-
"train/total_loss": 0.
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 30.0,
|
| 772 |
-
"eval_exact_match_accuracy": 0.
|
| 773 |
-
"eval_hamming_loss": 0.
|
| 774 |
-
"eval_loss": 0.
|
| 775 |
-
"eval_macro_f1": 0.
|
| 776 |
-
"eval_macro_precision": 0.
|
| 777 |
-
"eval_macro_recall": 0.
|
| 778 |
-
"eval_micro_f1": 0.
|
| 779 |
-
"eval_micro_precision": 0.
|
| 780 |
-
"eval_micro_recall": 0.
|
| 781 |
-
"eval_runtime": 0.
|
| 782 |
-
"eval_samples_per_second":
|
| 783 |
-
"eval_steps_per_second":
|
| 784 |
"step": 600
|
| 785 |
},
|
| 786 |
{
|
| 787 |
"epoch": 30.0,
|
| 788 |
"step": 600,
|
| 789 |
-
"train/classification_loss": 0.
|
| 790 |
-
"train/contrastive_loss": 0.
|
| 791 |
-
"train/negative_loss": 0.
|
| 792 |
"train/num_negatives": 38,
|
| 793 |
"train/num_positives": 18,
|
| 794 |
-
"train/positive_loss": 0.
|
| 795 |
-
"train/total_loss": 0.
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 30.0,
|
| 799 |
"step": 600,
|
| 800 |
-
"train/classification_loss": 0.
|
| 801 |
-
"train/contrastive_loss":
|
| 802 |
-
"train/negative_loss": 0.
|
| 803 |
"train/num_negatives": 48,
|
| 804 |
"train/num_positives": 8,
|
| 805 |
-
"train/positive_loss": 0.
|
| 806 |
-
"train/total_loss": 0.
|
| 807 |
},
|
| 808 |
{
|
| 809 |
"epoch": 32.5,
|
| 810 |
-
"grad_norm":
|
| 811 |
-
"learning_rate":
|
| 812 |
-
"loss": 0.
|
| 813 |
"step": 650
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 32.5,
|
| 817 |
"step": 650,
|
| 818 |
-
"train/classification_loss": 0.
|
| 819 |
-
"train/contrastive_loss": 0.
|
| 820 |
-
"train/negative_loss": 0.
|
| 821 |
"train/num_negatives": 36,
|
| 822 |
"train/num_positives": 18,
|
| 823 |
-
"train/positive_loss": 0.
|
| 824 |
-
"train/total_loss": 0.
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 32.5,
|
| 828 |
"step": 650,
|
| 829 |
-
"train/classification_loss": 0.
|
| 830 |
-
"train/contrastive_loss": 1.
|
| 831 |
-
"train/negative_loss": 1.
|
| 832 |
"train/num_negatives": 50,
|
| 833 |
"train/num_positives": 6,
|
| 834 |
-
"train/positive_loss": 0.
|
| 835 |
-
"train/total_loss": 0.
|
| 836 |
},
|
| 837 |
{
|
| 838 |
"epoch": 35.0,
|
| 839 |
-
"grad_norm":
|
| 840 |
-
"learning_rate":
|
| 841 |
-
"loss": 0.
|
| 842 |
"step": 700
|
| 843 |
},
|
| 844 |
{
|
| 845 |
"epoch": 35.0,
|
| 846 |
"step": 700,
|
| 847 |
-
"train/classification_loss": 0.
|
| 848 |
-
"train/contrastive_loss":
|
| 849 |
-
"train/negative_loss":
|
| 850 |
"train/num_negatives": 44,
|
| 851 |
"train/num_positives": 12,
|
| 852 |
-
"train/positive_loss":
|
| 853 |
-
"train/total_loss": 0.
|
| 854 |
},
|
| 855 |
{
|
| 856 |
"epoch": 35.0,
|
| 857 |
"step": 700,
|
| 858 |
-
"train/classification_loss": 0.
|
| 859 |
-
"train/contrastive_loss":
|
| 860 |
-
"train/negative_loss":
|
| 861 |
"train/num_negatives": 44,
|
| 862 |
"train/num_positives": 12,
|
| 863 |
-
"train/positive_loss": 0.
|
| 864 |
-
"train/total_loss":
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 35.0,
|
| 868 |
"step": 700,
|
| 869 |
-
"train/classification_loss": 0.
|
| 870 |
-
"train/contrastive_loss": 4.
|
| 871 |
-
"train/negative_loss": 3.
|
| 872 |
"train/num_negatives": 52,
|
| 873 |
"train/num_positives": 4,
|
| 874 |
-
"train/positive_loss": 1.
|
| 875 |
-
"train/total_loss": 1.
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 35.0,
|
| 879 |
"step": 700,
|
| 880 |
-
"train/classification_loss": 0.
|
| 881 |
-
"train/contrastive_loss": 2.
|
| 882 |
-
"train/negative_loss": 1.
|
| 883 |
"train/num_negatives": 48,
|
| 884 |
"train/num_positives": 8,
|
| 885 |
-
"train/positive_loss": 0.
|
| 886 |
-
"train/total_loss": 0.
|
| 887 |
},
|
| 888 |
{
|
| 889 |
"epoch": 35.0,
|
| 890 |
"step": 700,
|
| 891 |
-
"train/classification_loss": 0.
|
| 892 |
-
"train/contrastive_loss": 0.
|
| 893 |
-
"train/negative_loss": 0.
|
| 894 |
"train/num_negatives": 46,
|
| 895 |
"train/num_positives": 8,
|
| 896 |
-
"train/positive_loss": 0.
|
| 897 |
-
"train/total_loss": 0.
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 35.0,
|
| 901 |
"eval_exact_match_accuracy": 0.25,
|
| 902 |
-
"eval_hamming_loss": 0.
|
| 903 |
-
"eval_loss": 0.
|
| 904 |
-
"eval_macro_f1": 0.
|
| 905 |
-
"eval_macro_precision": 0.
|
| 906 |
-
"eval_macro_recall": 0.
|
| 907 |
-
"eval_micro_f1": 0.
|
| 908 |
-
"eval_micro_precision": 0.
|
| 909 |
-
"eval_micro_recall": 0.
|
| 910 |
-
"eval_runtime": 0.
|
| 911 |
-
"eval_samples_per_second":
|
| 912 |
-
"eval_steps_per_second": 24.
|
| 913 |
"step": 700
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 35.0,
|
| 917 |
"step": 700,
|
| 918 |
-
"train/classification_loss": 0.
|
| 919 |
-
"train/contrastive_loss": 0.
|
| 920 |
-
"train/negative_loss": 0.
|
| 921 |
"train/num_negatives": 40,
|
| 922 |
"train/num_positives": 16,
|
| 923 |
-
"train/positive_loss": 0.
|
| 924 |
-
"train/total_loss": 0.
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 35.0,
|
| 928 |
"step": 700,
|
| 929 |
-
"train/classification_loss": 0.
|
| 930 |
-
"train/contrastive_loss":
|
| 931 |
-
"train/negative_loss": 0.
|
| 932 |
"train/num_negatives": 40,
|
| 933 |
"train/num_positives": 16,
|
| 934 |
-
"train/positive_loss": 0.
|
| 935 |
-
"train/total_loss": 0.
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 37.5,
|
| 939 |
-
"grad_norm":
|
| 940 |
-
"learning_rate":
|
| 941 |
-
"loss": 0.
|
| 942 |
"step": 750
|
| 943 |
},
|
| 944 |
{
|
| 945 |
"epoch": 37.5,
|
| 946 |
"step": 750,
|
| 947 |
-
"train/classification_loss": 0.
|
| 948 |
-
"train/contrastive_loss": 0.
|
| 949 |
-
"train/negative_loss": 0.
|
| 950 |
"train/num_negatives": 48,
|
| 951 |
"train/num_positives": 8,
|
| 952 |
-
"train/positive_loss":
|
| 953 |
-
"train/total_loss": 0.
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 37.5,
|
| 957 |
"step": 750,
|
| 958 |
-
"train/classification_loss": 0.
|
| 959 |
-
"train/contrastive_loss": 0.
|
| 960 |
-
"train/negative_loss": 0.
|
| 961 |
"train/num_negatives": 46,
|
| 962 |
"train/num_positives": 10,
|
| 963 |
-
"train/positive_loss": 0.
|
| 964 |
-
"train/total_loss": 0.
|
| 965 |
},
|
| 966 |
{
|
| 967 |
"epoch": 40.0,
|
| 968 |
-
"grad_norm": 8.
|
| 969 |
-
"learning_rate":
|
| 970 |
-
"loss": 0.
|
| 971 |
"step": 800
|
| 972 |
},
|
| 973 |
{
|
| 974 |
"epoch": 40.0,
|
| 975 |
"step": 800,
|
| 976 |
-
"train/classification_loss": 0.
|
| 977 |
-
"train/contrastive_loss":
|
| 978 |
-
"train/negative_loss":
|
| 979 |
"train/num_negatives": 44,
|
| 980 |
"train/num_positives": 12,
|
| 981 |
-
"train/positive_loss":
|
| 982 |
-
"train/total_loss": 0.
|
| 983 |
},
|
| 984 |
{
|
| 985 |
"epoch": 40.0,
|
| 986 |
"step": 800,
|
| 987 |
-
"train/classification_loss": 0.
|
| 988 |
-
"train/contrastive_loss": 3.
|
| 989 |
-
"train/negative_loss": 3.
|
| 990 |
"train/num_negatives": 44,
|
| 991 |
"train/num_positives": 12,
|
| 992 |
-
"train/positive_loss": 0.
|
| 993 |
-
"train/total_loss":
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 40.0,
|
| 997 |
"step": 800,
|
| 998 |
-
"train/classification_loss": 0.
|
| 999 |
-
"train/contrastive_loss":
|
| 1000 |
-
"train/negative_loss": 2.
|
| 1001 |
"train/num_negatives": 52,
|
| 1002 |
"train/num_positives": 4,
|
| 1003 |
-
"train/positive_loss":
|
| 1004 |
-
"train/total_loss": 1.
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
"epoch": 40.0,
|
| 1008 |
"step": 800,
|
| 1009 |
-
"train/classification_loss": 0.
|
| 1010 |
-
"train/contrastive_loss":
|
| 1011 |
-
"train/negative_loss":
|
| 1012 |
"train/num_negatives": 48,
|
| 1013 |
"train/num_positives": 8,
|
| 1014 |
-
"train/positive_loss": 0.
|
| 1015 |
-
"train/total_loss": 0.
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
"epoch": 40.0,
|
| 1019 |
"step": 800,
|
| 1020 |
-
"train/classification_loss": 0.
|
| 1021 |
-
"train/contrastive_loss":
|
| 1022 |
-
"train/negative_loss": 0.
|
| 1023 |
"train/num_negatives": 46,
|
| 1024 |
"train/num_positives": 8,
|
| 1025 |
-
"train/positive_loss": 0.
|
| 1026 |
-
"train/total_loss": 0.
|
| 1027 |
},
|
| 1028 |
{
|
| 1029 |
"epoch": 40.0,
|
| 1030 |
-
"eval_exact_match_accuracy": 0.
|
| 1031 |
"eval_hamming_loss": 0.06470588235294118,
|
| 1032 |
-
"eval_loss": 0.
|
| 1033 |
-
"eval_macro_f1": 0.
|
| 1034 |
-
"eval_macro_precision": 0.
|
| 1035 |
-
"eval_macro_recall": 0.
|
| 1036 |
"eval_micro_f1": 0.42105263157894735,
|
| 1037 |
"eval_micro_precision": 0.8421052631578947,
|
| 1038 |
"eval_micro_recall": 0.2807017543859649,
|
| 1039 |
-
"eval_runtime": 0.
|
| 1040 |
-
"eval_samples_per_second":
|
| 1041 |
-
"eval_steps_per_second":
|
| 1042 |
"step": 800
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
"epoch": 40.0,
|
| 1046 |
"step": 800,
|
| 1047 |
-
"train/classification_loss": 0.
|
| 1048 |
-
"train/contrastive_loss": 0.
|
| 1049 |
-
"train/negative_loss": 0.
|
| 1050 |
"train/num_negatives": 44,
|
| 1051 |
"train/num_positives": 12,
|
| 1052 |
-
"train/positive_loss":
|
| 1053 |
-
"train/total_loss": 0.
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
"epoch": 40.0,
|
| 1057 |
"step": 800,
|
| 1058 |
-
"train/classification_loss": 0.
|
| 1059 |
-
"train/contrastive_loss": 0.
|
| 1060 |
-
"train/negative_loss": 0.
|
| 1061 |
"train/num_negatives": 30,
|
| 1062 |
"train/num_positives": 26,
|
| 1063 |
-
"train/positive_loss":
|
| 1064 |
-
"train/total_loss": 0.
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
"epoch": 42.5,
|
| 1068 |
-
"grad_norm":
|
| 1069 |
-
"learning_rate":
|
| 1070 |
-
"loss": 0.
|
| 1071 |
"step": 850
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
"epoch": 42.5,
|
| 1075 |
"step": 850,
|
| 1076 |
-
"train/classification_loss": 0.
|
| 1077 |
-
"train/contrastive_loss": 0.
|
| 1078 |
-
"train/negative_loss": 0.
|
| 1079 |
"train/num_negatives": 40,
|
| 1080 |
"train/num_positives": 16,
|
| 1081 |
-
"train/positive_loss": 0.
|
| 1082 |
-
"train/total_loss": 0.
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
"epoch": 42.5,
|
| 1086 |
"step": 850,
|
| 1087 |
-
"train/classification_loss": 0.
|
| 1088 |
-
"train/contrastive_loss": 0.
|
| 1089 |
-
"train/negative_loss": 0.
|
| 1090 |
"train/num_negatives": 40,
|
| 1091 |
"train/num_positives": 16,
|
| 1092 |
-
"train/positive_loss":
|
| 1093 |
-
"train/total_loss": 0.
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
"epoch": 45.0,
|
| 1097 |
-
"grad_norm":
|
| 1098 |
-
"learning_rate":
|
| 1099 |
-
"loss": 0.
|
| 1100 |
"step": 900
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 45.0,
|
| 1104 |
"step": 900,
|
| 1105 |
-
"train/classification_loss": 0.
|
| 1106 |
-
"train/contrastive_loss":
|
| 1107 |
-
"train/negative_loss":
|
| 1108 |
"train/num_negatives": 44,
|
| 1109 |
"train/num_positives": 12,
|
| 1110 |
-
"train/positive_loss":
|
| 1111 |
-
"train/total_loss":
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
"epoch": 45.0,
|
| 1115 |
"step": 900,
|
| 1116 |
-
"train/classification_loss": 0.
|
| 1117 |
-
"train/contrastive_loss": 3.
|
| 1118 |
-
"train/negative_loss": 3.
|
| 1119 |
"train/num_negatives": 44,
|
| 1120 |
"train/num_positives": 12,
|
| 1121 |
-
"train/positive_loss": 0.
|
| 1122 |
-
"train/total_loss": 0.
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
"epoch": 45.0,
|
| 1126 |
"step": 900,
|
| 1127 |
-
"train/classification_loss": 0.
|
| 1128 |
-
"train/contrastive_loss": 4.
|
| 1129 |
-
"train/negative_loss":
|
| 1130 |
"train/num_negatives": 52,
|
| 1131 |
"train/num_positives": 4,
|
| 1132 |
-
"train/positive_loss": 1.
|
| 1133 |
-
"train/total_loss": 1.
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
"epoch": 45.0,
|
| 1137 |
"step": 900,
|
| 1138 |
-
"train/classification_loss": 0.
|
| 1139 |
-
"train/contrastive_loss":
|
| 1140 |
-
"train/negative_loss":
|
| 1141 |
"train/num_negatives": 48,
|
| 1142 |
"train/num_positives": 8,
|
| 1143 |
-
"train/positive_loss": 0.
|
| 1144 |
-
"train/total_loss": 0.
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
"epoch": 45.0,
|
| 1148 |
"step": 900,
|
| 1149 |
-
"train/classification_loss": 0.
|
| 1150 |
-
"train/contrastive_loss":
|
| 1151 |
-
"train/negative_loss": 0.
|
| 1152 |
"train/num_negatives": 46,
|
| 1153 |
"train/num_positives": 8,
|
| 1154 |
-
"train/positive_loss":
|
| 1155 |
-
"train/total_loss": 0.
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
"epoch": 45.0,
|
| 1159 |
-
"eval_exact_match_accuracy": 0.
|
| 1160 |
-
"eval_hamming_loss": 0.
|
| 1161 |
-
"eval_loss": 0.
|
| 1162 |
-
"eval_macro_f1": 0.
|
| 1163 |
-
"eval_macro_precision": 0.
|
| 1164 |
-
"eval_macro_recall": 0.
|
| 1165 |
-
"eval_micro_f1": 0.
|
| 1166 |
-
"eval_micro_precision": 0.
|
| 1167 |
"eval_micro_recall": 0.2807017543859649,
|
| 1168 |
-
"eval_runtime": 0.
|
| 1169 |
-
"eval_samples_per_second":
|
| 1170 |
-
"eval_steps_per_second":
|
| 1171 |
"step": 900
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 45.0,
|
| 1175 |
"step": 900,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1176 |
"total_flos": 0.0,
|
| 1177 |
-
"train_loss": 0.
|
| 1178 |
-
"train_runtime":
|
| 1179 |
-
"train_samples_per_second":
|
| 1180 |
-
"train_steps_per_second":
|
| 1181 |
}
|
| 1182 |
],
|
| 1183 |
"logging_steps": 50,
|
| 1184 |
-
"max_steps":
|
| 1185 |
"num_input_tokens_seen": 0,
|
| 1186 |
-
"num_train_epochs":
|
| 1187 |
-
"save_steps":
|
| 1188 |
"stateful_callbacks": {
|
| 1189 |
"EarlyStoppingCallback": {
|
| 1190 |
"args": {
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 700,
|
| 3 |
"best_metric": 0.42105263157894735,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 50.0,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 12 |
{
|
| 13 |
"epoch": 0,
|
| 14 |
"step": 0,
|
| 15 |
+
"train/classification_loss": 0.6896063685417175,
|
| 16 |
+
"train/contrastive_loss": 9.115899085998535,
|
| 17 |
+
"train/negative_loss": 9.115803718566895,
|
| 18 |
"train/num_negatives": 46,
|
| 19 |
"train/num_positives": 10,
|
| 20 |
+
"train/positive_loss": 9.557702287565917e-05,
|
| 21 |
+
"train/total_loss": 2.5127861499786377
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0,
|
| 25 |
"step": 0,
|
| 26 |
+
"train/classification_loss": 0.7036023736000061,
|
| 27 |
+
"train/contrastive_loss": 8.648970603942871,
|
| 28 |
+
"train/negative_loss": 8.648796081542969,
|
| 29 |
"train/num_negatives": 46,
|
| 30 |
"train/num_positives": 10,
|
| 31 |
+
"train/positive_loss": 0.0001749610819388181,
|
| 32 |
+
"train/total_loss": 2.433396577835083
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 2.5,
|
| 36 |
+
"grad_norm": 18.205059051513672,
|
| 37 |
+
"learning_rate": 9.4e-06,
|
| 38 |
+
"loss": 3.6074,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.5,
|
| 43 |
"step": 50,
|
| 44 |
+
"train/classification_loss": 0.6570730805397034,
|
| 45 |
+
"train/contrastive_loss": 1.7841553688049316,
|
| 46 |
+
"train/negative_loss": 1.5581305027008057,
|
| 47 |
"train/num_negatives": 38,
|
| 48 |
"train/num_positives": 18,
|
| 49 |
+
"train/positive_loss": 0.22602489590644836,
|
| 50 |
+
"train/total_loss": 1.013904094696045
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 2.5,
|
| 54 |
"step": 50,
|
| 55 |
+
"train/classification_loss": 0.6499872207641602,
|
| 56 |
+
"train/contrastive_loss": 1.9876362085342407,
|
| 57 |
+
"train/negative_loss": 1.9320745468139648,
|
| 58 |
"train/num_negatives": 50,
|
| 59 |
"train/num_positives": 6,
|
| 60 |
+
"train/positive_loss": 0.05556164309382439,
|
| 61 |
+
"train/total_loss": 1.0475144386291504
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 5.0,
|
| 65 |
+
"grad_norm": 13.659893035888672,
|
| 66 |
+
"learning_rate": 1.94e-05,
|
| 67 |
+
"loss": 1.796,
|
| 68 |
"step": 100
|
| 69 |
},
|
| 70 |
{
|
| 71 |
"epoch": 5.0,
|
| 72 |
"step": 100,
|
| 73 |
+
"train/classification_loss": 0.49852314591407776,
|
| 74 |
+
"train/contrastive_loss": 2.1901540756225586,
|
| 75 |
+
"train/negative_loss": 1.469272494316101,
|
| 76 |
"train/num_negatives": 44,
|
| 77 |
"train/num_positives": 12,
|
| 78 |
+
"train/positive_loss": 0.7208815813064575,
|
| 79 |
+
"train/total_loss": 0.936553955078125
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 5.0,
|
| 83 |
"step": 100,
|
| 84 |
+
"train/classification_loss": 0.5119871497154236,
|
| 85 |
+
"train/contrastive_loss": 2.697293996810913,
|
| 86 |
+
"train/negative_loss": 2.695932388305664,
|
| 87 |
"train/num_negatives": 44,
|
| 88 |
"train/num_positives": 12,
|
| 89 |
+
"train/positive_loss": 0.0013614993076771498,
|
| 90 |
+
"train/total_loss": 1.0514459609985352
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 5.0,
|
| 94 |
"step": 100,
|
| 95 |
+
"train/classification_loss": 0.49341386556625366,
|
| 96 |
+
"train/contrastive_loss": 1.4812039136886597,
|
| 97 |
+
"train/negative_loss": 1.4763706922531128,
|
| 98 |
"train/num_negatives": 52,
|
| 99 |
"train/num_positives": 4,
|
| 100 |
+
"train/positive_loss": 0.004833280108869076,
|
| 101 |
+
"train/total_loss": 0.7896546125411987
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 5.0,
|
| 105 |
"step": 100,
|
| 106 |
+
"train/classification_loss": 0.49600034952163696,
|
| 107 |
+
"train/contrastive_loss": 3.470733165740967,
|
| 108 |
+
"train/negative_loss": 2.635651111602783,
|
| 109 |
"train/num_negatives": 48,
|
| 110 |
"train/num_positives": 8,
|
| 111 |
+
"train/positive_loss": 0.835081934928894,
|
| 112 |
+
"train/total_loss": 1.1901469230651855
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 5.0,
|
| 116 |
"step": 100,
|
| 117 |
+
"train/classification_loss": 0.49302685260772705,
|
| 118 |
+
"train/contrastive_loss": 1.5160608291625977,
|
| 119 |
+
"train/negative_loss": 1.3313394784927368,
|
| 120 |
"train/num_negatives": 46,
|
| 121 |
"train/num_positives": 8,
|
| 122 |
+
"train/positive_loss": 0.18472139537334442,
|
| 123 |
+
"train/total_loss": 0.7962390184402466
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 5.0,
|
| 127 |
"eval_exact_match_accuracy": 0.0,
|
| 128 |
+
"eval_hamming_loss": 0.08235294117647059,
|
| 129 |
+
"eval_loss": 0.9528080821037292,
|
| 130 |
+
"eval_macro_f1": 0.00980392156862745,
|
| 131 |
+
"eval_macro_precision": 0.058823529411764705,
|
| 132 |
+
"eval_macro_recall": 0.005347593582887701,
|
| 133 |
+
"eval_micro_f1": 0.034482758620689655,
|
| 134 |
+
"eval_micro_precision": 1.0,
|
| 135 |
+
"eval_micro_recall": 0.017543859649122806,
|
| 136 |
+
"eval_runtime": 0.2063,
|
| 137 |
+
"eval_samples_per_second": 193.875,
|
| 138 |
+
"eval_steps_per_second": 24.234,
|
| 139 |
"step": 100
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 5.0,
|
| 143 |
"step": 100,
|
| 144 |
+
"train/classification_loss": 0.5103878378868103,
|
| 145 |
+
"train/contrastive_loss": 1.4061696529388428,
|
| 146 |
+
"train/negative_loss": 0.860652506351471,
|
| 147 |
"train/num_negatives": 42,
|
| 148 |
"train/num_positives": 12,
|
| 149 |
+
"train/positive_loss": 0.5455171465873718,
|
| 150 |
+
"train/total_loss": 0.7916218042373657
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 5.0,
|
| 154 |
"step": 100,
|
| 155 |
+
"train/classification_loss": 0.49409669637680054,
|
| 156 |
+
"train/contrastive_loss": 0.6720238327980042,
|
| 157 |
+
"train/negative_loss": 0.5605456233024597,
|
| 158 |
"train/num_negatives": 40,
|
| 159 |
"train/num_positives": 16,
|
| 160 |
+
"train/positive_loss": 0.11147819459438324,
|
| 161 |
+
"train/total_loss": 0.6285014748573303
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 7.5,
|
| 165 |
+
"grad_norm": 14.211859703063965,
|
| 166 |
+
"learning_rate": 1.9905050505050507e-05,
|
| 167 |
+
"loss": 1.3708,
|
| 168 |
"step": 150
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 7.5,
|
| 172 |
"step": 150,
|
| 173 |
+
"train/classification_loss": 0.40184715390205383,
|
| 174 |
+
"train/contrastive_loss": 1.165258765220642,
|
| 175 |
+
"train/negative_loss": 0.698477566242218,
|
| 176 |
"train/num_negatives": 44,
|
| 177 |
"train/num_positives": 8,
|
| 178 |
+
"train/positive_loss": 0.4667811989784241,
|
| 179 |
+
"train/total_loss": 0.6348989009857178
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 7.5,
|
| 183 |
"step": 150,
|
| 184 |
+
"train/classification_loss": 0.3879987299442291,
|
| 185 |
+
"train/contrastive_loss": 0.8239786028862,
|
| 186 |
+
"train/negative_loss": 0.7370874285697937,
|
| 187 |
"train/num_negatives": 44,
|
| 188 |
"train/num_positives": 10,
|
| 189 |
+
"train/positive_loss": 0.08689115941524506,
|
| 190 |
+
"train/total_loss": 0.5527944564819336
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 10.0,
|
| 194 |
+
"grad_norm": 25.8126277923584,
|
| 195 |
+
"learning_rate": 1.9804040404040406e-05,
|
| 196 |
+
"loss": 1.142,
|
| 197 |
"step": 200
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 10.0,
|
| 201 |
"step": 200,
|
| 202 |
+
"train/classification_loss": 0.2806307375431061,
|
| 203 |
+
"train/contrastive_loss": 2.156310558319092,
|
| 204 |
+
"train/negative_loss": 2.146817445755005,
|
| 205 |
"train/num_negatives": 44,
|
| 206 |
"train/num_positives": 12,
|
| 207 |
+
"train/positive_loss": 0.009493212215602398,
|
| 208 |
+
"train/total_loss": 0.71189284324646
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"epoch": 10.0,
|
| 212 |
"step": 200,
|
| 213 |
+
"train/classification_loss": 0.3437887132167816,
|
| 214 |
+
"train/contrastive_loss": 3.3835480213165283,
|
| 215 |
+
"train/negative_loss": 3.3831517696380615,
|
| 216 |
"train/num_negatives": 44,
|
| 217 |
"train/num_positives": 12,
|
| 218 |
+
"train/positive_loss": 0.00039613869739696383,
|
| 219 |
+
"train/total_loss": 1.020498275756836
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 10.0,
|
| 223 |
"step": 200,
|
| 224 |
+
"train/classification_loss": 0.31499814987182617,
|
| 225 |
+
"train/contrastive_loss": 2.7768659591674805,
|
| 226 |
+
"train/negative_loss": 2.7635276317596436,
|
| 227 |
"train/num_negatives": 52,
|
| 228 |
"train/num_positives": 4,
|
| 229 |
+
"train/positive_loss": 0.013338344171643257,
|
| 230 |
+
"train/total_loss": 0.8703713417053223
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 10.0,
|
| 234 |
"step": 200,
|
| 235 |
+
"train/classification_loss": 0.3134444057941437,
|
| 236 |
+
"train/contrastive_loss": 3.8373122215270996,
|
| 237 |
+
"train/negative_loss": 2.7304084300994873,
|
| 238 |
"train/num_negatives": 48,
|
| 239 |
"train/num_positives": 8,
|
| 240 |
+
"train/positive_loss": 1.1069039106369019,
|
| 241 |
+
"train/total_loss": 1.080906867980957
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"epoch": 10.0,
|
| 245 |
"step": 200,
|
| 246 |
+
"train/classification_loss": 0.3265349864959717,
|
| 247 |
+
"train/contrastive_loss": 1.5294461250305176,
|
| 248 |
+
"train/negative_loss": 1.4643787145614624,
|
| 249 |
"train/num_negatives": 46,
|
| 250 |
"train/num_positives": 8,
|
| 251 |
+
"train/positive_loss": 0.06506740301847458,
|
| 252 |
+
"train/total_loss": 0.6324242353439331
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 10.0,
|
| 256 |
"eval_exact_match_accuracy": 0.0,
|
| 257 |
+
"eval_hamming_loss": 0.08235294117647059,
|
| 258 |
+
"eval_loss": 0.8632186651229858,
|
| 259 |
+
"eval_macro_f1": 0.00980392156862745,
|
| 260 |
+
"eval_macro_precision": 0.058823529411764705,
|
| 261 |
+
"eval_macro_recall": 0.005347593582887701,
|
| 262 |
+
"eval_micro_f1": 0.034482758620689655,
|
| 263 |
+
"eval_micro_precision": 1.0,
|
| 264 |
+
"eval_micro_recall": 0.017543859649122806,
|
| 265 |
+
"eval_runtime": 0.2085,
|
| 266 |
+
"eval_samples_per_second": 191.856,
|
| 267 |
+
"eval_steps_per_second": 23.982,
|
| 268 |
"step": 200
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 10.0,
|
| 272 |
"step": 200,
|
| 273 |
+
"train/classification_loss": 0.3289062976837158,
|
| 274 |
+
"train/contrastive_loss": 1.158113956451416,
|
| 275 |
+
"train/negative_loss": 0.9101917147636414,
|
| 276 |
"train/num_negatives": 42,
|
| 277 |
"train/num_positives": 14,
|
| 278 |
+
"train/positive_loss": 0.24792218208312988,
|
| 279 |
+
"train/total_loss": 0.5605291128158569
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 10.0,
|
| 283 |
"step": 200,
|
| 284 |
+
"train/classification_loss": 0.28807589411735535,
|
| 285 |
+
"train/contrastive_loss": 1.227806568145752,
|
| 286 |
+
"train/negative_loss": 1.2268885374069214,
|
| 287 |
"train/num_negatives": 42,
|
| 288 |
"train/num_positives": 14,
|
| 289 |
+
"train/positive_loss": 0.0009180314373224974,
|
| 290 |
+
"train/total_loss": 0.5336372256278992
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 12.5,
|
| 294 |
+
"grad_norm": 7.029545307159424,
|
| 295 |
+
"learning_rate": 1.9703030303030302e-05,
|
| 296 |
+
"loss": 1.0178,
|
| 297 |
"step": 250
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 12.5,
|
| 301 |
"step": 250,
|
| 302 |
+
"train/classification_loss": 0.24251216650009155,
|
| 303 |
+
"train/contrastive_loss": 1.3313517570495605,
|
| 304 |
+
"train/negative_loss": 1.1213274002075195,
|
| 305 |
"train/num_negatives": 44,
|
| 306 |
"train/num_positives": 12,
|
| 307 |
+
"train/positive_loss": 0.21002434194087982,
|
| 308 |
+
"train/total_loss": 0.5087825059890747
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 12.5,
|
| 312 |
"step": 250,
|
| 313 |
+
"train/classification_loss": 0.2935709059238434,
|
| 314 |
+
"train/contrastive_loss": 0.40547823905944824,
|
| 315 |
+
"train/negative_loss": 0.3695138692855835,
|
| 316 |
"train/num_negatives": 42,
|
| 317 |
"train/num_positives": 8,
|
| 318 |
+
"train/positive_loss": 0.03596435487270355,
|
| 319 |
+
"train/total_loss": 0.37466657161712646
|
| 320 |
},
|
| 321 |
{
|
| 322 |
"epoch": 15.0,
|
| 323 |
+
"grad_norm": 10.424208641052246,
|
| 324 |
+
"learning_rate": 1.9602020202020205e-05,
|
| 325 |
+
"loss": 0.8805,
|
| 326 |
"step": 300
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 15.0,
|
| 330 |
"step": 300,
|
| 331 |
+
"train/classification_loss": 0.22187121212482452,
|
| 332 |
+
"train/contrastive_loss": 2.8647470474243164,
|
| 333 |
+
"train/negative_loss": 2.7984812259674072,
|
| 334 |
"train/num_negatives": 44,
|
| 335 |
"train/num_positives": 12,
|
| 336 |
+
"train/positive_loss": 0.0662657767534256,
|
| 337 |
+
"train/total_loss": 0.7948206067085266
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 15.0,
|
| 341 |
"step": 300,
|
| 342 |
+
"train/classification_loss": 0.2955247759819031,
|
| 343 |
+
"train/contrastive_loss": 4.269000053405762,
|
| 344 |
+
"train/negative_loss": 4.268981456756592,
|
| 345 |
"train/num_negatives": 44,
|
| 346 |
"train/num_positives": 12,
|
| 347 |
+
"train/positive_loss": 1.871622953331098e-05,
|
| 348 |
+
"train/total_loss": 1.1493247747421265
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 15.0,
|
| 352 |
"step": 300,
|
| 353 |
+
"train/classification_loss": 0.26941171288490295,
|
| 354 |
+
"train/contrastive_loss": 4.073124885559082,
|
| 355 |
+
"train/negative_loss": 2.919600486755371,
|
| 356 |
"train/num_negatives": 52,
|
| 357 |
"train/num_positives": 4,
|
| 358 |
+
"train/positive_loss": 1.1535242795944214,
|
| 359 |
+
"train/total_loss": 1.0840367078781128
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 15.0,
|
| 363 |
"step": 300,
|
| 364 |
+
"train/classification_loss": 0.2642236351966858,
|
| 365 |
+
"train/contrastive_loss": 5.231786727905273,
|
| 366 |
+
"train/negative_loss": 4.430575370788574,
|
| 367 |
"train/num_negatives": 48,
|
| 368 |
"train/num_positives": 8,
|
| 369 |
+
"train/positive_loss": 0.8012115359306335,
|
| 370 |
+
"train/total_loss": 1.3105809688568115
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 15.0,
|
| 374 |
"step": 300,
|
| 375 |
+
"train/classification_loss": 0.2807493805885315,
|
| 376 |
+
"train/contrastive_loss": 1.4644469022750854,
|
| 377 |
+
"train/negative_loss": 1.4178317785263062,
|
| 378 |
"train/num_negatives": 46,
|
| 379 |
"train/num_positives": 8,
|
| 380 |
+
"train/positive_loss": 0.04661515727639198,
|
| 381 |
+
"train/total_loss": 0.5736387968063354
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 15.0,
|
| 385 |
"eval_exact_match_accuracy": 0.05,
|
| 386 |
+
"eval_hamming_loss": 0.07794117647058824,
|
| 387 |
+
"eval_loss": 0.9824804067611694,
|
| 388 |
+
"eval_macro_f1": 0.03460207612456748,
|
| 389 |
+
"eval_macro_precision": 0.049019607843137254,
|
| 390 |
"eval_macro_recall": 0.026737967914438502,
|
| 391 |
+
"eval_micro_f1": 0.15873015873015872,
|
| 392 |
+
"eval_micro_precision": 0.8333333333333334,
|
| 393 |
"eval_micro_recall": 0.08771929824561403,
|
| 394 |
+
"eval_runtime": 0.2153,
|
| 395 |
+
"eval_samples_per_second": 185.81,
|
| 396 |
+
"eval_steps_per_second": 23.226,
|
| 397 |
"step": 300
|
| 398 |
},
|
| 399 |
{
|
| 400 |
"epoch": 15.0,
|
| 401 |
"step": 300,
|
| 402 |
+
"train/classification_loss": 0.24673022329807281,
|
| 403 |
+
"train/contrastive_loss": 0.4545953869819641,
|
| 404 |
+
"train/negative_loss": 0.443155974149704,
|
| 405 |
"train/num_negatives": 32,
|
| 406 |
"train/num_positives": 20,
|
| 407 |
+
"train/positive_loss": 0.011439427733421326,
|
| 408 |
+
"train/total_loss": 0.33764928579330444
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 15.0,
|
| 412 |
"step": 300,
|
| 413 |
+
"train/classification_loss": 0.2520601153373718,
|
| 414 |
+
"train/contrastive_loss": 1.4950696229934692,
|
| 415 |
+
"train/negative_loss": 1.429970145225525,
|
| 416 |
"train/num_negatives": 46,
|
| 417 |
"train/num_positives": 10,
|
| 418 |
+
"train/positive_loss": 0.06509946286678314,
|
| 419 |
+
"train/total_loss": 0.5510740280151367
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 17.5,
|
| 423 |
+
"grad_norm": 9.645092964172363,
|
| 424 |
+
"learning_rate": 1.95010101010101e-05,
|
| 425 |
+
"loss": 0.7785,
|
| 426 |
"step": 350
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 17.5,
|
| 430 |
"step": 350,
|
| 431 |
+
"train/classification_loss": 0.2225627303123474,
|
| 432 |
+
"train/contrastive_loss": 0.3660619258880615,
|
| 433 |
+
"train/negative_loss": 0.36337095499038696,
|
| 434 |
"train/num_negatives": 42,
|
| 435 |
"train/num_positives": 14,
|
| 436 |
+
"train/positive_loss": 0.0026909795124083757,
|
| 437 |
+
"train/total_loss": 0.2957751154899597
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 17.5,
|
| 441 |
"step": 350,
|
| 442 |
+
"train/classification_loss": 0.2583811581134796,
|
| 443 |
+
"train/contrastive_loss": 0.25796303153038025,
|
| 444 |
+
"train/negative_loss": 0.25655531883239746,
|
| 445 |
"train/num_negatives": 44,
|
| 446 |
"train/num_positives": 6,
|
| 447 |
+
"train/positive_loss": 0.0014077159576117992,
|
| 448 |
+
"train/total_loss": 0.3099737763404846
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 20.0,
|
| 452 |
+
"grad_norm": 5.90708065032959,
|
| 453 |
+
"learning_rate": 1.94e-05,
|
| 454 |
+
"loss": 0.7442,
|
| 455 |
"step": 400
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 20.0,
|
| 459 |
"step": 400,
|
| 460 |
+
"train/classification_loss": 0.1851058006286621,
|
| 461 |
+
"train/contrastive_loss": 2.0512137413024902,
|
| 462 |
+
"train/negative_loss": 1.7761945724487305,
|
| 463 |
"train/num_negatives": 44,
|
| 464 |
"train/num_positives": 12,
|
| 465 |
+
"train/positive_loss": 0.27501922845840454,
|
| 466 |
+
"train/total_loss": 0.595348596572876
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 20.0,
|
| 470 |
"step": 400,
|
| 471 |
+
"train/classification_loss": 0.2684085965156555,
|
| 472 |
+
"train/contrastive_loss": 3.0632779598236084,
|
| 473 |
+
"train/negative_loss": 3.05769419670105,
|
| 474 |
"train/num_negatives": 44,
|
| 475 |
"train/num_positives": 12,
|
| 476 |
+
"train/positive_loss": 0.0055836960673332214,
|
| 477 |
+
"train/total_loss": 0.8810641765594482
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 20.0,
|
| 481 |
"step": 400,
|
| 482 |
+
"train/classification_loss": 0.25380370020866394,
|
| 483 |
+
"train/contrastive_loss": 4.407430648803711,
|
| 484 |
+
"train/negative_loss": 3.715952157974243,
|
| 485 |
"train/num_negatives": 52,
|
| 486 |
"train/num_positives": 4,
|
| 487 |
+
"train/positive_loss": 0.6914786100387573,
|
| 488 |
+
"train/total_loss": 1.1352897882461548
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 20.0,
|
| 492 |
"step": 400,
|
| 493 |
+
"train/classification_loss": 0.23351122438907623,
|
| 494 |
+
"train/contrastive_loss": 2.3350398540496826,
|
| 495 |
+
"train/negative_loss": 1.7342523336410522,
|
| 496 |
"train/num_negatives": 48,
|
| 497 |
"train/num_positives": 8,
|
| 498 |
+
"train/positive_loss": 0.6007875800132751,
|
| 499 |
+
"train/total_loss": 0.7005192041397095
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 20.0,
|
| 503 |
"step": 400,
|
| 504 |
+
"train/classification_loss": 0.2567104995250702,
|
| 505 |
+
"train/contrastive_loss": 1.2914104461669922,
|
| 506 |
+
"train/negative_loss": 1.1075626611709595,
|
| 507 |
"train/num_negatives": 46,
|
| 508 |
"train/num_positives": 8,
|
| 509 |
+
"train/positive_loss": 0.18384775519371033,
|
| 510 |
+
"train/total_loss": 0.5149925947189331
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 20.0,
|
| 514 |
"eval_exact_match_accuracy": 0.1,
|
| 515 |
"eval_hamming_loss": 0.075,
|
| 516 |
+
"eval_loss": 0.7654428482055664,
|
| 517 |
+
"eval_macro_f1": 0.0803921568627451,
|
| 518 |
+
"eval_macro_precision": 0.1045751633986928,
|
| 519 |
+
"eval_macro_recall": 0.06684491978609625,
|
| 520 |
+
"eval_micro_f1": 0.23880597014925373,
|
| 521 |
+
"eval_micro_precision": 0.8,
|
| 522 |
+
"eval_micro_recall": 0.14035087719298245,
|
| 523 |
+
"eval_runtime": 0.2071,
|
| 524 |
+
"eval_samples_per_second": 193.116,
|
| 525 |
+
"eval_steps_per_second": 24.139,
|
| 526 |
"step": 400
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 20.0,
|
| 530 |
"step": 400,
|
| 531 |
+
"train/classification_loss": 0.23103763163089752,
|
| 532 |
+
"train/contrastive_loss": 0.9073829650878906,
|
| 533 |
+
"train/negative_loss": 0.5319927930831909,
|
| 534 |
"train/num_negatives": 42,
|
| 535 |
"train/num_positives": 10,
|
| 536 |
+
"train/positive_loss": 0.3753901422023773,
|
| 537 |
+
"train/total_loss": 0.41251420974731445
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 20.0,
|
| 541 |
"step": 400,
|
| 542 |
+
"train/classification_loss": 0.2296096831560135,
|
| 543 |
+
"train/contrastive_loss": 0.5736224055290222,
|
| 544 |
+
"train/negative_loss": 0.573104977607727,
|
| 545 |
"train/num_negatives": 52,
|
| 546 |
"train/num_positives": 4,
|
| 547 |
+
"train/positive_loss": 0.0005174159305170178,
|
| 548 |
+
"train/total_loss": 0.3443341553211212
|
| 549 |
},
|
| 550 |
{
|
| 551 |
"epoch": 22.5,
|
| 552 |
+
"grad_norm": 9.208708763122559,
|
| 553 |
+
"learning_rate": 1.92989898989899e-05,
|
| 554 |
+
"loss": 0.6661,
|
| 555 |
"step": 450
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 22.5,
|
| 559 |
"step": 450,
|
| 560 |
+
"train/classification_loss": 0.21061192452907562,
|
| 561 |
+
"train/contrastive_loss": 1.1704968214035034,
|
| 562 |
+
"train/negative_loss": 1.170390248298645,
|
| 563 |
"train/num_negatives": 50,
|
| 564 |
"train/num_positives": 4,
|
| 565 |
+
"train/positive_loss": 0.00010655130608938634,
|
| 566 |
+
"train/total_loss": 0.444711297750473
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 22.5,
|
| 570 |
"step": 450,
|
| 571 |
+
"train/classification_loss": 0.18421001732349396,
|
| 572 |
+
"train/contrastive_loss": 1.406832218170166,
|
| 573 |
+
"train/negative_loss": 0.8008975982666016,
|
| 574 |
"train/num_negatives": 50,
|
| 575 |
"train/num_positives": 6,
|
| 576 |
+
"train/positive_loss": 0.6059346795082092,
|
| 577 |
+
"train/total_loss": 0.4655764698982239
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 25.0,
|
| 581 |
+
"grad_norm": 9.534709930419922,
|
| 582 |
+
"learning_rate": 1.91979797979798e-05,
|
| 583 |
+
"loss": 0.6332,
|
| 584 |
"step": 500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 25.0,
|
| 588 |
"step": 500,
|
| 589 |
+
"train/classification_loss": 0.17407798767089844,
|
| 590 |
+
"train/contrastive_loss": 3.3778998851776123,
|
| 591 |
+
"train/negative_loss": 3.208540678024292,
|
| 592 |
"train/num_negatives": 44,
|
| 593 |
"train/num_positives": 12,
|
| 594 |
+
"train/positive_loss": 0.16935913264751434,
|
| 595 |
+
"train/total_loss": 0.8496579527854919
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 25.0,
|
| 599 |
"step": 500,
|
| 600 |
+
"train/classification_loss": 0.25022628903388977,
|
| 601 |
+
"train/contrastive_loss": 2.501643180847168,
|
| 602 |
+
"train/negative_loss": 2.5011374950408936,
|
| 603 |
"train/num_negatives": 44,
|
| 604 |
"train/num_positives": 12,
|
| 605 |
+
"train/positive_loss": 0.0005057163070887327,
|
| 606 |
+
"train/total_loss": 0.7505549192428589
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 25.0,
|
| 610 |
"step": 500,
|
| 611 |
+
"train/classification_loss": 0.24351510405540466,
|
| 612 |
+
"train/contrastive_loss": 5.5227251052856445,
|
| 613 |
+
"train/negative_loss": 3.6524746417999268,
|
| 614 |
"train/num_negatives": 52,
|
| 615 |
"train/num_positives": 4,
|
| 616 |
+
"train/positive_loss": 1.8702502250671387,
|
| 617 |
+
"train/total_loss": 1.348060131072998
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 25.0,
|
| 621 |
"step": 500,
|
| 622 |
+
"train/classification_loss": 0.20675694942474365,
|
| 623 |
+
"train/contrastive_loss": 2.3911428451538086,
|
| 624 |
+
"train/negative_loss": 1.8613694906234741,
|
| 625 |
"train/num_negatives": 48,
|
| 626 |
"train/num_positives": 8,
|
| 627 |
+
"train/positive_loss": 0.5297732949256897,
|
| 628 |
+
"train/total_loss": 0.6849855184555054
|
| 629 |
},
|
| 630 |
{
|
| 631 |
"epoch": 25.0,
|
| 632 |
"step": 500,
|
| 633 |
+
"train/classification_loss": 0.24148425459861755,
|
| 634 |
+
"train/contrastive_loss": 1.3859646320343018,
|
| 635 |
+
"train/negative_loss": 0.7059023380279541,
|
| 636 |
"train/num_negatives": 46,
|
| 637 |
"train/num_positives": 8,
|
| 638 |
+
"train/positive_loss": 0.6800622940063477,
|
| 639 |
+
"train/total_loss": 0.5186771750450134
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 25.0,
|
| 643 |
"eval_exact_match_accuracy": 0.175,
|
| 644 |
+
"eval_hamming_loss": 0.07352941176470588,
|
| 645 |
+
"eval_loss": 0.8303871154785156,
|
| 646 |
+
"eval_macro_f1": 0.10802139037433156,
|
| 647 |
+
"eval_macro_precision": 0.14331550802139037,
|
| 648 |
+
"eval_macro_recall": 0.09037433155080213,
|
| 649 |
+
"eval_micro_f1": 0.32432432432432434,
|
| 650 |
+
"eval_micro_precision": 0.7058823529411765,
|
| 651 |
+
"eval_micro_recall": 0.21052631578947367,
|
| 652 |
+
"eval_runtime": 0.2078,
|
| 653 |
+
"eval_samples_per_second": 192.449,
|
| 654 |
+
"eval_steps_per_second": 24.056,
|
| 655 |
"step": 500
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 25.0,
|
| 659 |
"step": 500,
|
| 660 |
+
"train/classification_loss": 0.14580048620700836,
|
| 661 |
+
"train/contrastive_loss": 0.2984220087528229,
|
| 662 |
+
"train/negative_loss": 0.29449352622032166,
|
| 663 |
"train/num_negatives": 42,
|
| 664 |
"train/num_positives": 14,
|
| 665 |
+
"train/positive_loss": 0.003928476478904486,
|
| 666 |
+
"train/total_loss": 0.20548489689826965
|
| 667 |
},
|
| 668 |
{
|
| 669 |
"epoch": 25.0,
|
| 670 |
"step": 500,
|
| 671 |
+
"train/classification_loss": 0.23653770983219147,
|
| 672 |
+
"train/contrastive_loss": 0.7606168389320374,
|
| 673 |
+
"train/negative_loss": 0.737991988658905,
|
| 674 |
"train/num_negatives": 44,
|
| 675 |
"train/num_positives": 12,
|
| 676 |
+
"train/positive_loss": 0.022624850273132324,
|
| 677 |
+
"train/total_loss": 0.38866108655929565
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 27.5,
|
| 681 |
+
"grad_norm": 6.362896919250488,
|
| 682 |
+
"learning_rate": 1.90969696969697e-05,
|
| 683 |
+
"loss": 0.5843,
|
| 684 |
"step": 550
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 27.5,
|
| 688 |
"step": 550,
|
| 689 |
+
"train/classification_loss": 0.2056518942117691,
|
| 690 |
+
"train/contrastive_loss": 0.9691342711448669,
|
| 691 |
+
"train/negative_loss": 0.9691179394721985,
|
| 692 |
"train/num_negatives": 54,
|
| 693 |
"train/num_positives": 2,
|
| 694 |
+
"train/positive_loss": 1.633180545468349e-05,
|
| 695 |
+
"train/total_loss": 0.3994787335395813
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 27.5,
|
| 699 |
"step": 550,
|
| 700 |
+
"train/classification_loss": 0.22270001471042633,
|
| 701 |
+
"train/contrastive_loss": 1.5889058113098145,
|
| 702 |
+
"train/negative_loss": 1.251386046409607,
|
| 703 |
"train/num_negatives": 40,
|
| 704 |
"train/num_positives": 12,
|
| 705 |
+
"train/positive_loss": 0.3375198245048523,
|
| 706 |
+
"train/total_loss": 0.5404812097549438
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 30.0,
|
| 710 |
+
"grad_norm": 9.474464416503906,
|
| 711 |
+
"learning_rate": 1.8995959595959595e-05,
|
| 712 |
+
"loss": 0.5572,
|
| 713 |
"step": 600
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 30.0,
|
| 717 |
"step": 600,
|
| 718 |
+
"train/classification_loss": 0.16965512931346893,
|
| 719 |
+
"train/contrastive_loss": 1.8685212135314941,
|
| 720 |
+
"train/negative_loss": 1.4877853393554688,
|
| 721 |
"train/num_negatives": 44,
|
| 722 |
"train/num_positives": 12,
|
| 723 |
+
"train/positive_loss": 0.3807358741760254,
|
| 724 |
+
"train/total_loss": 0.5433593988418579
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 30.0,
|
| 728 |
"step": 600,
|
| 729 |
+
"train/classification_loss": 0.2326871156692505,
|
| 730 |
+
"train/contrastive_loss": 3.803422451019287,
|
| 731 |
+
"train/negative_loss": 3.799872875213623,
|
| 732 |
"train/num_negatives": 44,
|
| 733 |
"train/num_positives": 12,
|
| 734 |
+
"train/positive_loss": 0.003549614455550909,
|
| 735 |
+
"train/total_loss": 0.9933716058731079
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 30.0,
|
| 739 |
"step": 600,
|
| 740 |
+
"train/classification_loss": 0.21336901187896729,
|
| 741 |
+
"train/contrastive_loss": 2.9807915687561035,
|
| 742 |
+
"train/negative_loss": 2.322481155395508,
|
| 743 |
"train/num_negatives": 52,
|
| 744 |
"train/num_positives": 4,
|
| 745 |
+
"train/positive_loss": 0.6583104133605957,
|
| 746 |
+
"train/total_loss": 0.8095273375511169
|
| 747 |
},
|
| 748 |
{
|
| 749 |
"epoch": 30.0,
|
| 750 |
"step": 600,
|
| 751 |
+
"train/classification_loss": 0.21435928344726562,
|
| 752 |
+
"train/contrastive_loss": 4.913402080535889,
|
| 753 |
+
"train/negative_loss": 4.489163875579834,
|
| 754 |
"train/num_negatives": 48,
|
| 755 |
"train/num_positives": 8,
|
| 756 |
+
"train/positive_loss": 0.4242383539676666,
|
| 757 |
+
"train/total_loss": 1.1970397233963013
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 30.0,
|
| 761 |
"step": 600,
|
| 762 |
+
"train/classification_loss": 0.22257567942142487,
|
| 763 |
+
"train/contrastive_loss": 0.9281859993934631,
|
| 764 |
+
"train/negative_loss": 0.7518287301063538,
|
| 765 |
"train/num_negatives": 46,
|
| 766 |
"train/num_positives": 8,
|
| 767 |
+
"train/positive_loss": 0.17635725438594818,
|
| 768 |
+
"train/total_loss": 0.40821290016174316
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 30.0,
|
| 772 |
+
"eval_exact_match_accuracy": 0.225,
|
| 773 |
+
"eval_hamming_loss": 0.06764705882352941,
|
| 774 |
+
"eval_loss": 0.7903022170066833,
|
| 775 |
+
"eval_macro_f1": 0.11995386389850057,
|
| 776 |
+
"eval_macro_precision": 0.1596638655462185,
|
| 777 |
+
"eval_macro_recall": 0.09679144385026739,
|
| 778 |
+
"eval_micro_f1": 0.3611111111111111,
|
| 779 |
+
"eval_micro_precision": 0.8666666666666667,
|
| 780 |
+
"eval_micro_recall": 0.22807017543859648,
|
| 781 |
+
"eval_runtime": 0.2099,
|
| 782 |
+
"eval_samples_per_second": 190.603,
|
| 783 |
+
"eval_steps_per_second": 23.825,
|
| 784 |
"step": 600
|
| 785 |
},
|
| 786 |
{
|
| 787 |
"epoch": 30.0,
|
| 788 |
"step": 600,
|
| 789 |
+
"train/classification_loss": 0.14723831415176392,
|
| 790 |
+
"train/contrastive_loss": 0.6149296760559082,
|
| 791 |
+
"train/negative_loss": 0.4006352722644806,
|
| 792 |
"train/num_negatives": 38,
|
| 793 |
"train/num_positives": 18,
|
| 794 |
+
"train/positive_loss": 0.21429438889026642,
|
| 795 |
+
"train/total_loss": 0.2702242434024811
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 30.0,
|
| 799 |
"step": 600,
|
| 800 |
+
"train/classification_loss": 0.18867409229278564,
|
| 801 |
+
"train/contrastive_loss": 0.7789785265922546,
|
| 802 |
+
"train/negative_loss": 0.6380741000175476,
|
| 803 |
"train/num_negatives": 48,
|
| 804 |
"train/num_positives": 8,
|
| 805 |
+
"train/positive_loss": 0.14090441167354584,
|
| 806 |
+
"train/total_loss": 0.3444697856903076
|
| 807 |
},
|
| 808 |
{
|
| 809 |
"epoch": 32.5,
|
| 810 |
+
"grad_norm": 7.496462821960449,
|
| 811 |
+
"learning_rate": 1.8894949494949498e-05,
|
| 812 |
+
"loss": 0.5177,
|
| 813 |
"step": 650
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 32.5,
|
| 817 |
"step": 650,
|
| 818 |
+
"train/classification_loss": 0.1460404396057129,
|
| 819 |
+
"train/contrastive_loss": 0.8524841070175171,
|
| 820 |
+
"train/negative_loss": 0.623662531375885,
|
| 821 |
"train/num_negatives": 36,
|
| 822 |
"train/num_positives": 18,
|
| 823 |
+
"train/positive_loss": 0.22882159054279327,
|
| 824 |
+
"train/total_loss": 0.3165372610092163
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 32.5,
|
| 828 |
"step": 650,
|
| 829 |
+
"train/classification_loss": 0.18225882947444916,
|
| 830 |
+
"train/contrastive_loss": 1.4429905414581299,
|
| 831 |
+
"train/negative_loss": 1.4358431100845337,
|
| 832 |
"train/num_negatives": 50,
|
| 833 |
"train/num_positives": 6,
|
| 834 |
+
"train/positive_loss": 0.007147490046918392,
|
| 835 |
+
"train/total_loss": 0.4708569645881653
|
| 836 |
},
|
| 837 |
{
|
| 838 |
"epoch": 35.0,
|
| 839 |
+
"grad_norm": 6.565901756286621,
|
| 840 |
+
"learning_rate": 1.8793939393939394e-05,
|
| 841 |
+
"loss": 0.4788,
|
| 842 |
"step": 700
|
| 843 |
},
|
| 844 |
{
|
| 845 |
"epoch": 35.0,
|
| 846 |
"step": 700,
|
| 847 |
+
"train/classification_loss": 0.16156421601772308,
|
| 848 |
+
"train/contrastive_loss": 3.9083566665649414,
|
| 849 |
+
"train/negative_loss": 2.979792594909668,
|
| 850 |
"train/num_negatives": 44,
|
| 851 |
"train/num_positives": 12,
|
| 852 |
+
"train/positive_loss": 0.9285640716552734,
|
| 853 |
+
"train/total_loss": 0.9432355761528015
|
| 854 |
},
|
| 855 |
{
|
| 856 |
"epoch": 35.0,
|
| 857 |
"step": 700,
|
| 858 |
+
"train/classification_loss": 0.22515544295310974,
|
| 859 |
+
"train/contrastive_loss": 3.891486167907715,
|
| 860 |
+
"train/negative_loss": 3.8570120334625244,
|
| 861 |
"train/num_negatives": 44,
|
| 862 |
"train/num_positives": 12,
|
| 863 |
+
"train/positive_loss": 0.034474026411771774,
|
| 864 |
+
"train/total_loss": 1.0034526586532593
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 35.0,
|
| 868 |
"step": 700,
|
| 869 |
+
"train/classification_loss": 0.22171786427497864,
|
| 870 |
+
"train/contrastive_loss": 4.330800533294678,
|
| 871 |
+
"train/negative_loss": 3.1184730529785156,
|
| 872 |
"train/num_negatives": 52,
|
| 873 |
"train/num_positives": 4,
|
| 874 |
+
"train/positive_loss": 1.2123275995254517,
|
| 875 |
+
"train/total_loss": 1.0878779888153076
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 35.0,
|
| 879 |
"step": 700,
|
| 880 |
+
"train/classification_loss": 0.1916349083185196,
|
| 881 |
+
"train/contrastive_loss": 2.0170962810516357,
|
| 882 |
+
"train/negative_loss": 1.338474154472351,
|
| 883 |
"train/num_negatives": 48,
|
| 884 |
"train/num_positives": 8,
|
| 885 |
+
"train/positive_loss": 0.6786221861839294,
|
| 886 |
+
"train/total_loss": 0.5950541496276855
|
| 887 |
},
|
| 888 |
{
|
| 889 |
"epoch": 35.0,
|
| 890 |
"step": 700,
|
| 891 |
+
"train/classification_loss": 0.20504897832870483,
|
| 892 |
+
"train/contrastive_loss": 0.6244919896125793,
|
| 893 |
+
"train/negative_loss": 0.39054688811302185,
|
| 894 |
"train/num_negatives": 46,
|
| 895 |
"train/num_positives": 8,
|
| 896 |
+
"train/positive_loss": 0.2339451164007187,
|
| 897 |
+
"train/total_loss": 0.3299473822116852
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 35.0,
|
| 901 |
"eval_exact_match_accuracy": 0.25,
|
| 902 |
+
"eval_hamming_loss": 0.06470588235294118,
|
| 903 |
+
"eval_loss": 0.7919135689735413,
|
| 904 |
+
"eval_macro_f1": 0.14236282748393475,
|
| 905 |
+
"eval_macro_precision": 0.21512605042016808,
|
| 906 |
+
"eval_macro_recall": 0.11729055258467024,
|
| 907 |
+
"eval_micro_f1": 0.42105263157894735,
|
| 908 |
+
"eval_micro_precision": 0.8421052631578947,
|
| 909 |
+
"eval_micro_recall": 0.2807017543859649,
|
| 910 |
+
"eval_runtime": 0.2075,
|
| 911 |
+
"eval_samples_per_second": 192.814,
|
| 912 |
+
"eval_steps_per_second": 24.102,
|
| 913 |
"step": 700
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 35.0,
|
| 917 |
"step": 700,
|
| 918 |
+
"train/classification_loss": 0.14606674015522003,
|
| 919 |
+
"train/contrastive_loss": 0.6812934875488281,
|
| 920 |
+
"train/negative_loss": 0.45186591148376465,
|
| 921 |
"train/num_negatives": 40,
|
| 922 |
"train/num_positives": 16,
|
| 923 |
+
"train/positive_loss": 0.2294275462627411,
|
| 924 |
+
"train/total_loss": 0.2823254466056824
|
| 925 |
},
|
| 926 |
{
|
| 927 |
"epoch": 35.0,
|
| 928 |
"step": 700,
|
| 929 |
+
"train/classification_loss": 0.1429990828037262,
|
| 930 |
+
"train/contrastive_loss": 0.5248066782951355,
|
| 931 |
+
"train/negative_loss": 0.3964087963104248,
|
| 932 |
"train/num_negatives": 40,
|
| 933 |
"train/num_positives": 16,
|
| 934 |
+
"train/positive_loss": 0.1283978968858719,
|
| 935 |
+
"train/total_loss": 0.2479604184627533
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 37.5,
|
| 939 |
+
"grad_norm": 3.286250114440918,
|
| 940 |
+
"learning_rate": 1.8692929292929294e-05,
|
| 941 |
+
"loss": 0.4698,
|
| 942 |
"step": 750
|
| 943 |
},
|
| 944 |
{
|
| 945 |
"epoch": 37.5,
|
| 946 |
"step": 750,
|
| 947 |
+
"train/classification_loss": 0.15292038023471832,
|
| 948 |
+
"train/contrastive_loss": 0.8152571320533752,
|
| 949 |
+
"train/negative_loss": 0.8152315616607666,
|
| 950 |
"train/num_negatives": 48,
|
| 951 |
"train/num_positives": 8,
|
| 952 |
+
"train/positive_loss": 2.557095831434708e-05,
|
| 953 |
+
"train/total_loss": 0.3159717917442322
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 37.5,
|
| 957 |
"step": 750,
|
| 958 |
+
"train/classification_loss": 0.15736867487430573,
|
| 959 |
+
"train/contrastive_loss": 0.38875800371170044,
|
| 960 |
+
"train/negative_loss": 0.26881521940231323,
|
| 961 |
"train/num_negatives": 46,
|
| 962 |
"train/num_positives": 10,
|
| 963 |
+
"train/positive_loss": 0.11994278430938721,
|
| 964 |
+
"train/total_loss": 0.2351202666759491
|
| 965 |
},
|
| 966 |
{
|
| 967 |
"epoch": 40.0,
|
| 968 |
+
"grad_norm": 8.739147186279297,
|
| 969 |
+
"learning_rate": 1.8591919191919193e-05,
|
| 970 |
+
"loss": 0.418,
|
| 971 |
"step": 800
|
| 972 |
},
|
| 973 |
{
|
| 974 |
"epoch": 40.0,
|
| 975 |
"step": 800,
|
| 976 |
+
"train/classification_loss": 0.16104553639888763,
|
| 977 |
+
"train/contrastive_loss": 3.0951993465423584,
|
| 978 |
+
"train/negative_loss": 1.8000407218933105,
|
| 979 |
"train/num_negatives": 44,
|
| 980 |
"train/num_positives": 12,
|
| 981 |
+
"train/positive_loss": 1.2951586246490479,
|
| 982 |
+
"train/total_loss": 0.7800854444503784
|
| 983 |
},
|
| 984 |
{
|
| 985 |
"epoch": 40.0,
|
| 986 |
"step": 800,
|
| 987 |
+
"train/classification_loss": 0.22088098526000977,
|
| 988 |
+
"train/contrastive_loss": 3.92935848236084,
|
| 989 |
+
"train/negative_loss": 3.9107954502105713,
|
| 990 |
"train/num_negatives": 44,
|
| 991 |
"train/num_positives": 12,
|
| 992 |
+
"train/positive_loss": 0.018563104793429375,
|
| 993 |
+
"train/total_loss": 1.0067527294158936
|
| 994 |
},
|
| 995 |
{
|
| 996 |
"epoch": 40.0,
|
| 997 |
"step": 800,
|
| 998 |
+
"train/classification_loss": 0.20756281912326813,
|
| 999 |
+
"train/contrastive_loss": 5.228389739990234,
|
| 1000 |
+
"train/negative_loss": 2.9686391353607178,
|
| 1001 |
"train/num_negatives": 52,
|
| 1002 |
"train/num_positives": 4,
|
| 1003 |
+
"train/positive_loss": 2.2597508430480957,
|
| 1004 |
+
"train/total_loss": 1.2532408237457275
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
"epoch": 40.0,
|
| 1008 |
"step": 800,
|
| 1009 |
+
"train/classification_loss": 0.1891981065273285,
|
| 1010 |
+
"train/contrastive_loss": 1.6292307376861572,
|
| 1011 |
+
"train/negative_loss": 1.227111577987671,
|
| 1012 |
"train/num_negatives": 48,
|
| 1013 |
"train/num_positives": 8,
|
| 1014 |
+
"train/positive_loss": 0.40211910009384155,
|
| 1015 |
+
"train/total_loss": 0.5150442719459534
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
"epoch": 40.0,
|
| 1019 |
"step": 800,
|
| 1020 |
+
"train/classification_loss": 0.19581612944602966,
|
| 1021 |
+
"train/contrastive_loss": 0.958620548248291,
|
| 1022 |
+
"train/negative_loss": 0.6886929869651794,
|
| 1023 |
"train/num_negatives": 46,
|
| 1024 |
"train/num_positives": 8,
|
| 1025 |
+
"train/positive_loss": 0.2699275612831116,
|
| 1026 |
+
"train/total_loss": 0.38754022121429443
|
| 1027 |
},
|
| 1028 |
{
|
| 1029 |
"epoch": 40.0,
|
| 1030 |
+
"eval_exact_match_accuracy": 0.2,
|
| 1031 |
"eval_hamming_loss": 0.06470588235294118,
|
| 1032 |
+
"eval_loss": 0.7885327339172363,
|
| 1033 |
+
"eval_macro_f1": 0.18102240896358546,
|
| 1034 |
+
"eval_macro_precision": 0.3300653594771242,
|
| 1035 |
+
"eval_macro_recall": 0.1354723707664884,
|
| 1036 |
"eval_micro_f1": 0.42105263157894735,
|
| 1037 |
"eval_micro_precision": 0.8421052631578947,
|
| 1038 |
"eval_micro_recall": 0.2807017543859649,
|
| 1039 |
+
"eval_runtime": 0.2094,
|
| 1040 |
+
"eval_samples_per_second": 191.056,
|
| 1041 |
+
"eval_steps_per_second": 23.882,
|
| 1042 |
"step": 800
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
"epoch": 40.0,
|
| 1046 |
"step": 800,
|
| 1047 |
+
"train/classification_loss": 0.12362033128738403,
|
| 1048 |
+
"train/contrastive_loss": 0.3485001027584076,
|
| 1049 |
+
"train/negative_loss": 0.3484821319580078,
|
| 1050 |
"train/num_negatives": 44,
|
| 1051 |
"train/num_positives": 12,
|
| 1052 |
+
"train/positive_loss": 1.7980994016397744e-05,
|
| 1053 |
+
"train/total_loss": 0.1933203637599945
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
"epoch": 40.0,
|
| 1057 |
"step": 800,
|
| 1058 |
+
"train/classification_loss": 0.09559500962495804,
|
| 1059 |
+
"train/contrastive_loss": 0.22876113653182983,
|
| 1060 |
+
"train/negative_loss": 0.2287520468235016,
|
| 1061 |
"train/num_negatives": 30,
|
| 1062 |
"train/num_positives": 26,
|
| 1063 |
+
"train/positive_loss": 9.087464604817796e-06,
|
| 1064 |
+
"train/total_loss": 0.1413472294807434
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
"epoch": 42.5,
|
| 1068 |
+
"grad_norm": 3.746946334838867,
|
| 1069 |
+
"learning_rate": 1.8490909090909093e-05,
|
| 1070 |
+
"loss": 0.3927,
|
| 1071 |
"step": 850
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
"epoch": 42.5,
|
| 1075 |
"step": 850,
|
| 1076 |
+
"train/classification_loss": 0.09173674136400223,
|
| 1077 |
+
"train/contrastive_loss": 0.13494296371936798,
|
| 1078 |
+
"train/negative_loss": 0.13212107121944427,
|
| 1079 |
"train/num_negatives": 40,
|
| 1080 |
"train/num_positives": 16,
|
| 1081 |
+
"train/positive_loss": 0.002821887144818902,
|
| 1082 |
+
"train/total_loss": 0.11872533708810806
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
"epoch": 42.5,
|
| 1086 |
"step": 850,
|
| 1087 |
+
"train/classification_loss": 0.12884384393692017,
|
| 1088 |
+
"train/contrastive_loss": 0.07590682804584503,
|
| 1089 |
+
"train/negative_loss": 0.07587877660989761,
|
| 1090 |
"train/num_negatives": 40,
|
| 1091 |
"train/num_positives": 16,
|
| 1092 |
+
"train/positive_loss": 2.8053931600879878e-05,
|
| 1093 |
+
"train/total_loss": 0.14402520656585693
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
"epoch": 45.0,
|
| 1097 |
+
"grad_norm": 2.8717594146728516,
|
| 1098 |
+
"learning_rate": 1.8389898989898992e-05,
|
| 1099 |
+
"loss": 0.3975,
|
| 1100 |
"step": 900
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 45.0,
|
| 1104 |
"step": 900,
|
| 1105 |
+
"train/classification_loss": 0.16052784025669098,
|
| 1106 |
+
"train/contrastive_loss": 4.337796688079834,
|
| 1107 |
+
"train/negative_loss": 3.431094169616699,
|
| 1108 |
"train/num_negatives": 44,
|
| 1109 |
"train/num_positives": 12,
|
| 1110 |
+
"train/positive_loss": 0.9067023992538452,
|
| 1111 |
+
"train/total_loss": 1.0280872583389282
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
"epoch": 45.0,
|
| 1115 |
"step": 900,
|
| 1116 |
+
"train/classification_loss": 0.2030077874660492,
|
| 1117 |
+
"train/contrastive_loss": 3.2337257862091064,
|
| 1118 |
+
"train/negative_loss": 3.2324299812316895,
|
| 1119 |
"train/num_negatives": 44,
|
| 1120 |
"train/num_positives": 12,
|
| 1121 |
+
"train/positive_loss": 0.001295788330025971,
|
| 1122 |
+
"train/total_loss": 0.8497529029846191
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
"epoch": 45.0,
|
| 1126 |
"step": 900,
|
| 1127 |
+
"train/classification_loss": 0.2113606035709381,
|
| 1128 |
+
"train/contrastive_loss": 4.936408996582031,
|
| 1129 |
+
"train/negative_loss": 3.0232598781585693,
|
| 1130 |
"train/num_negatives": 52,
|
| 1131 |
"train/num_positives": 4,
|
| 1132 |
+
"train/positive_loss": 1.913149356842041,
|
| 1133 |
+
"train/total_loss": 1.198642373085022
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
"epoch": 45.0,
|
| 1137 |
"step": 900,
|
| 1138 |
+
"train/classification_loss": 0.17921508848667145,
|
| 1139 |
+
"train/contrastive_loss": 2.6765177249908447,
|
| 1140 |
+
"train/negative_loss": 1.9601576328277588,
|
| 1141 |
"train/num_negatives": 48,
|
| 1142 |
"train/num_positives": 8,
|
| 1143 |
+
"train/positive_loss": 0.7163600921630859,
|
| 1144 |
+
"train/total_loss": 0.7145186066627502
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
"epoch": 45.0,
|
| 1148 |
"step": 900,
|
| 1149 |
+
"train/classification_loss": 0.18929758667945862,
|
| 1150 |
+
"train/contrastive_loss": 0.7077760696411133,
|
| 1151 |
+
"train/negative_loss": 0.6388620138168335,
|
| 1152 |
"train/num_negatives": 46,
|
| 1153 |
"train/num_positives": 8,
|
| 1154 |
+
"train/positive_loss": 0.06891404092311859,
|
| 1155 |
+
"train/total_loss": 0.33085280656814575
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
"epoch": 45.0,
|
| 1159 |
+
"eval_exact_match_accuracy": 0.225,
|
| 1160 |
+
"eval_hamming_loss": 0.06911764705882353,
|
| 1161 |
+
"eval_loss": 0.8243707418441772,
|
| 1162 |
+
"eval_macro_f1": 0.15536244800950683,
|
| 1163 |
+
"eval_macro_precision": 0.2290552584670232,
|
| 1164 |
+
"eval_macro_recall": 0.12611408199643495,
|
| 1165 |
+
"eval_micro_f1": 0.4050632911392405,
|
| 1166 |
+
"eval_micro_precision": 0.7272727272727273,
|
| 1167 |
"eval_micro_recall": 0.2807017543859649,
|
| 1168 |
+
"eval_runtime": 0.2089,
|
| 1169 |
+
"eval_samples_per_second": 191.499,
|
| 1170 |
+
"eval_steps_per_second": 23.937,
|
| 1171 |
"step": 900
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 45.0,
|
| 1175 |
"step": 900,
|
| 1176 |
+
"train/classification_loss": 0.11641532182693481,
|
| 1177 |
+
"train/contrastive_loss": 1.0102875232696533,
|
| 1178 |
+
"train/negative_loss": 1.0102651119232178,
|
| 1179 |
+
"train/num_negatives": 54,
|
| 1180 |
+
"train/num_positives": 2,
|
| 1181 |
+
"train/positive_loss": 2.2411597456084564e-05,
|
| 1182 |
+
"train/total_loss": 0.31847283244132996
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 45.0,
|
| 1186 |
+
"step": 900,
|
| 1187 |
+
"train/classification_loss": 0.11128353327512741,
|
| 1188 |
+
"train/contrastive_loss": 0.24437068402767181,
|
| 1189 |
+
"train/negative_loss": 0.24322275817394257,
|
| 1190 |
+
"train/num_negatives": 46,
|
| 1191 |
+
"train/num_positives": 10,
|
| 1192 |
+
"train/positive_loss": 0.0011479274835437536,
|
| 1193 |
+
"train/total_loss": 0.16015766561031342
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 47.5,
|
| 1197 |
+
"grad_norm": 2.6908442974090576,
|
| 1198 |
+
"learning_rate": 1.8288888888888888e-05,
|
| 1199 |
+
"loss": 0.353,
|
| 1200 |
+
"step": 950
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 47.5,
|
| 1204 |
+
"step": 950,
|
| 1205 |
+
"train/classification_loss": 0.10373832285404205,
|
| 1206 |
+
"train/contrastive_loss": 0.37015679478645325,
|
| 1207 |
+
"train/negative_loss": 0.3693355917930603,
|
| 1208 |
+
"train/num_negatives": 46,
|
| 1209 |
+
"train/num_positives": 8,
|
| 1210 |
+
"train/positive_loss": 0.000821194495074451,
|
| 1211 |
+
"train/total_loss": 0.17776969075202942
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 47.5,
|
| 1215 |
+
"step": 950,
|
| 1216 |
+
"train/classification_loss": 0.1416180431842804,
|
| 1217 |
+
"train/contrastive_loss": 0.16073152422904968,
|
| 1218 |
+
"train/negative_loss": 0.15973255038261414,
|
| 1219 |
+
"train/num_negatives": 36,
|
| 1220 |
+
"train/num_positives": 16,
|
| 1221 |
+
"train/positive_loss": 0.0009989773388952017,
|
| 1222 |
+
"train/total_loss": 0.17376434803009033
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 50.0,
|
| 1226 |
+
"grad_norm": 5.927994251251221,
|
| 1227 |
+
"learning_rate": 1.818787878787879e-05,
|
| 1228 |
+
"loss": 0.3431,
|
| 1229 |
+
"step": 1000
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 50.0,
|
| 1233 |
+
"step": 1000,
|
| 1234 |
+
"train/classification_loss": 0.16533198952674866,
|
| 1235 |
+
"train/contrastive_loss": 3.354642868041992,
|
| 1236 |
+
"train/negative_loss": 2.9447507858276367,
|
| 1237 |
+
"train/num_negatives": 44,
|
| 1238 |
+
"train/num_positives": 12,
|
| 1239 |
+
"train/positive_loss": 0.409892201423645,
|
| 1240 |
+
"train/total_loss": 0.8362605571746826
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 50.0,
|
| 1244 |
+
"step": 1000,
|
| 1245 |
+
"train/classification_loss": 0.1996230185031891,
|
| 1246 |
+
"train/contrastive_loss": 2.7300727367401123,
|
| 1247 |
+
"train/negative_loss": 2.6814076900482178,
|
| 1248 |
+
"train/num_negatives": 44,
|
| 1249 |
+
"train/num_positives": 12,
|
| 1250 |
+
"train/positive_loss": 0.04866510629653931,
|
| 1251 |
+
"train/total_loss": 0.7456375360488892
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 50.0,
|
| 1255 |
+
"step": 1000,
|
| 1256 |
+
"train/classification_loss": 0.22982969880104065,
|
| 1257 |
+
"train/contrastive_loss": 5.4518961906433105,
|
| 1258 |
+
"train/negative_loss": 3.708037853240967,
|
| 1259 |
+
"train/num_negatives": 52,
|
| 1260 |
+
"train/num_positives": 4,
|
| 1261 |
+
"train/positive_loss": 1.7438582181930542,
|
| 1262 |
+
"train/total_loss": 1.3202089071273804
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 50.0,
|
| 1266 |
+
"step": 1000,
|
| 1267 |
+
"train/classification_loss": 0.19913451373577118,
|
| 1268 |
+
"train/contrastive_loss": 2.374002456665039,
|
| 1269 |
+
"train/negative_loss": 1.768825888633728,
|
| 1270 |
+
"train/num_negatives": 48,
|
| 1271 |
+
"train/num_positives": 8,
|
| 1272 |
+
"train/positive_loss": 0.6051765084266663,
|
| 1273 |
+
"train/total_loss": 0.6739349961280823
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 50.0,
|
| 1277 |
+
"step": 1000,
|
| 1278 |
+
"train/classification_loss": 0.1891087144613266,
|
| 1279 |
+
"train/contrastive_loss": 1.0433825254440308,
|
| 1280 |
+
"train/negative_loss": 0.5633068680763245,
|
| 1281 |
+
"train/num_negatives": 46,
|
| 1282 |
+
"train/num_positives": 8,
|
| 1283 |
+
"train/positive_loss": 0.4800756871700287,
|
| 1284 |
+
"train/total_loss": 0.3977852165699005
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 50.0,
|
| 1288 |
+
"eval_exact_match_accuracy": 0.225,
|
| 1289 |
+
"eval_hamming_loss": 0.06911764705882353,
|
| 1290 |
+
"eval_loss": 0.7947654128074646,
|
| 1291 |
+
"eval_macro_f1": 0.1929738562091503,
|
| 1292 |
+
"eval_macro_precision": 0.2908496732026144,
|
| 1293 |
+
"eval_macro_recall": 0.15017825311942956,
|
| 1294 |
+
"eval_micro_f1": 0.41975308641975306,
|
| 1295 |
+
"eval_micro_precision": 0.7083333333333334,
|
| 1296 |
+
"eval_micro_recall": 0.2982456140350877,
|
| 1297 |
+
"eval_runtime": 0.2051,
|
| 1298 |
+
"eval_samples_per_second": 194.982,
|
| 1299 |
+
"eval_steps_per_second": 24.373,
|
| 1300 |
+
"step": 1000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 50.0,
|
| 1304 |
+
"step": 1000,
|
| 1305 |
"total_flos": 0.0,
|
| 1306 |
+
"train_loss": 0.8574352493286133,
|
| 1307 |
+
"train_runtime": 257.7927,
|
| 1308 |
+
"train_samples_per_second": 612.896,
|
| 1309 |
+
"train_steps_per_second": 38.791
|
| 1310 |
}
|
| 1311 |
],
|
| 1312 |
"logging_steps": 50,
|
| 1313 |
+
"max_steps": 10000,
|
| 1314 |
"num_input_tokens_seen": 0,
|
| 1315 |
+
"num_train_epochs": 500,
|
| 1316 |
+
"save_steps": 10000,
|
| 1317 |
"stateful_callbacks": {
|
| 1318 |
"EarlyStoppingCallback": {
|
| 1319 |
"args": {
|