Model save
Browse files- README.md +1 -1
- all_results.json +5 -5
- model-00001-of-00004.safetensors +1 -1
- model-00002-of-00004.safetensors +1 -1
- model-00003-of-00004.safetensors +1 -1
- model-00004-of-00004.safetensors +1 -1
- train_results.json +5 -5
- trainer_state.json +1035 -1063
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -35,7 +35,7 @@ The following hyperparameters were used during training:
|
|
| 35 |
- learning_rate: 1e-06
|
| 36 |
- train_batch_size: 2
|
| 37 |
- eval_batch_size: 8
|
| 38 |
-
- seed:
|
| 39 |
- distributed_type: multi-GPU
|
| 40 |
- num_devices: 8
|
| 41 |
- gradient_accumulation_steps: 8
|
|
|
|
| 35 |
- learning_rate: 1e-06
|
| 36 |
- train_batch_size: 2
|
| 37 |
- eval_batch_size: 8
|
| 38 |
+
- seed: 1
|
| 39 |
- distributed_type: multi-GPU
|
| 40 |
- num_devices: 8
|
| 41 |
- gradient_accumulation_steps: 8
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.39703809490243847,
|
| 4 |
+
"train_runtime": 12665.7954,
|
| 5 |
+
"train_samples": 61134,
|
| 6 |
+
"train_samples_per_second": 9.653,
|
| 7 |
+
"train_steps_per_second": 0.075
|
| 8 |
}
|
model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4976698672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1aae0d51dec8f1db0c4185db4064e9e7cfa0f57226a7bdd1336e77d73fe07784
|
| 3 |
size 4976698672
|
model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999802720
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c74342a40e020ecc68e1877af3bfbc8f6458eacd08a2befc496f6b74d4cc12e8
|
| 3 |
size 4999802720
|
model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4915916176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dfcd78d697ab08df9806100c6023e029071e435e0071465482e099bcb80a107e
|
| 3 |
size 4915916176
|
model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1168138808
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:152de9df35d2cf2f89f72f30f0fcc0c733b787726e60d1a6e2659edcad51784a
|
| 3 |
size 1168138808
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.39703809490243847,
|
| 4 |
+
"train_runtime": 12665.7954,
|
| 5 |
+
"train_samples": 61134,
|
| 6 |
+
"train_samples_per_second": 9.653,
|
| 7 |
+
"train_steps_per_second": 0.075
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1383 +1,1355 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 1.
|
| 5 |
"eval_steps": 10000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
-
"learning_rate": 1.
|
| 14 |
-
"logits/chosen": 0.
|
| 15 |
-
"logits/rejected": 0.
|
| 16 |
-
"logps/chosen": -
|
| 17 |
-
"logps/rejected": -
|
| 18 |
-
"loss": 0.
|
| 19 |
-
"rewards/accuracies": 0.
|
| 20 |
-
"rewards/chosen": 0.
|
| 21 |
-
"rewards/margins": 0.
|
| 22 |
-
"rewards/rejected":
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.04,
|
| 27 |
-
"learning_rate": 2.
|
| 28 |
-
"logits/chosen": 0.
|
| 29 |
-
"logits/rejected": 0.
|
| 30 |
-
"logps/chosen": -
|
| 31 |
-
"logps/rejected": -
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"rewards/accuracies": 0.
|
| 34 |
-
"rewards/chosen":
|
| 35 |
-
"rewards/margins": 0.
|
| 36 |
-
"rewards/rejected": -0.
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.06,
|
| 41 |
-
"learning_rate": 3.
|
| 42 |
-
"logits/chosen":
|
| 43 |
-
"logits/rejected": 0.
|
| 44 |
-
"logps/chosen": -
|
| 45 |
-
"logps/rejected": -
|
| 46 |
-
"loss": 0.
|
| 47 |
-
"rewards/accuracies": 0.
|
| 48 |
-
"rewards/chosen": 0.
|
| 49 |
-
"rewards/margins": 0.
|
| 50 |
-
"rewards/rejected": -0.
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.08,
|
| 55 |
-
"learning_rate": 4.
|
| 56 |
-
"logits/chosen": 0.
|
| 57 |
-
"logits/rejected": 0.
|
| 58 |
-
"logps/chosen": -
|
| 59 |
-
"logps/rejected": -
|
| 60 |
-
"loss": 0.
|
| 61 |
-
"rewards/accuracies": 0.
|
| 62 |
-
"rewards/chosen": 0.
|
| 63 |
-
"rewards/margins": 0.
|
| 64 |
-
"rewards/rejected":
|
| 65 |
"step": 40
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.1,
|
| 69 |
-
"learning_rate": 5.
|
| 70 |
-
"logits/chosen": 0.
|
| 71 |
-
"logits/rejected": 0.
|
| 72 |
-
"logps/chosen": -
|
| 73 |
-
"logps/rejected": -
|
| 74 |
-
"loss": 0.
|
| 75 |
-
"rewards/accuracies": 0.
|
| 76 |
-
"rewards/chosen": 0.
|
| 77 |
-
"rewards/margins": 0.
|
| 78 |
-
"rewards/rejected":
|
| 79 |
"step": 50
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"learning_rate": 6.
|
| 84 |
-
"logits/chosen": 0.
|
| 85 |
-
"logits/rejected": 0.
|
| 86 |
-
"logps/chosen": -
|
| 87 |
-
"logps/rejected": -
|
| 88 |
-
"loss": 0.
|
| 89 |
-
"rewards/accuracies": 0.
|
| 90 |
-
"rewards/chosen":
|
| 91 |
-
"rewards/margins": 0.
|
| 92 |
-
"rewards/rejected": -0.
|
| 93 |
"step": 60
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"learning_rate": 7.
|
| 98 |
-
"logits/chosen": 0.
|
| 99 |
-
"logits/rejected": 0.
|
| 100 |
-
"logps/chosen": -
|
| 101 |
-
"logps/rejected": -
|
| 102 |
-
"loss": 0.
|
| 103 |
-
"rewards/accuracies": 0.
|
| 104 |
-
"rewards/chosen":
|
| 105 |
-
"rewards/margins": 0.
|
| 106 |
-
"rewards/rejected":
|
| 107 |
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"learning_rate": 8.
|
| 112 |
-
"logits/chosen": 0.
|
| 113 |
-
"logits/rejected": 0.
|
| 114 |
-
"logps/chosen": -
|
| 115 |
-
"logps/rejected": -
|
| 116 |
-
"loss": 0.
|
| 117 |
-
"rewards/accuracies": 0.
|
| 118 |
-
"rewards/chosen":
|
| 119 |
-
"rewards/margins": 0.
|
| 120 |
-
"rewards/rejected": -0.
|
| 121 |
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"learning_rate": 9.
|
| 126 |
-
"logits/chosen": 0.
|
| 127 |
-
"logits/rejected": 0.
|
| 128 |
-
"logps/chosen": -
|
| 129 |
-
"logps/rejected": -
|
| 130 |
-
"loss": 0.
|
| 131 |
-
"rewards/accuracies": 0.
|
| 132 |
-
"rewards/chosen":
|
| 133 |
-
"rewards/margins": 0.
|
| 134 |
-
"rewards/rejected": -0.
|
| 135 |
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"learning_rate": 9.
|
| 140 |
-
"logits/chosen":
|
| 141 |
-
"logits/rejected": 0.
|
| 142 |
-
"logps/chosen": -
|
| 143 |
-
"logps/rejected": -
|
| 144 |
-
"loss": 0.
|
| 145 |
"rewards/accuracies": 0.643750011920929,
|
| 146 |
-
"rewards/chosen": 0.
|
| 147 |
-
"rewards/margins": 0.
|
| 148 |
-
"rewards/rejected": -0.
|
| 149 |
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.23,
|
| 153 |
-
"learning_rate": 9.
|
| 154 |
-
"logits/chosen":
|
| 155 |
-
"logits/rejected": 0.
|
| 156 |
-
"logps/chosen": -
|
| 157 |
-
"logps/rejected": -
|
| 158 |
-
"loss": 0.
|
| 159 |
-
"rewards/accuracies": 0.
|
| 160 |
-
"rewards/chosen": 0.
|
| 161 |
-
"rewards/margins": 0.
|
| 162 |
-
"rewards/rejected": -0.
|
| 163 |
"step": 110
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.25,
|
| 167 |
-
"learning_rate": 9.
|
| 168 |
-
"logits/chosen":
|
| 169 |
-
"logits/rejected": 0.
|
| 170 |
-
"logps/chosen": -
|
| 171 |
-
"logps/rejected": -
|
| 172 |
-
"loss": 0.
|
| 173 |
-
"rewards/accuracies": 0.
|
| 174 |
-
"rewards/chosen": -0.
|
| 175 |
-
"rewards/margins": 0.
|
| 176 |
-
"rewards/rejected": -0.
|
| 177 |
"step": 120
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.27,
|
| 181 |
-
"learning_rate": 9.
|
| 182 |
-
"logits/chosen": 0.
|
| 183 |
-
"logits/rejected": 0.
|
| 184 |
-
"logps/chosen": -
|
| 185 |
-
"logps/rejected": -
|
| 186 |
-
"loss": 0.
|
| 187 |
-
"rewards/accuracies": 0.
|
| 188 |
-
"rewards/chosen": 0.
|
| 189 |
-
"rewards/margins": 0.
|
| 190 |
-
"rewards/rejected": -
|
| 191 |
"step": 130
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.29,
|
| 195 |
-
"learning_rate": 9.
|
| 196 |
-
"logits/chosen": 0.
|
| 197 |
-
"logits/rejected": 0.
|
| 198 |
-
"logps/chosen": -
|
| 199 |
-
"logps/rejected": -
|
| 200 |
-
"loss": 0.
|
| 201 |
-
"rewards/accuracies": 0.
|
| 202 |
-
"rewards/chosen": -
|
| 203 |
-
"rewards/margins":
|
| 204 |
-
"rewards/rejected": -
|
| 205 |
"step": 140
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.31,
|
| 209 |
-
"learning_rate": 9.
|
| 210 |
-
"logits/chosen": 0.
|
| 211 |
-
"logits/rejected": 0.
|
| 212 |
-
"logps/chosen": -
|
| 213 |
-
"logps/rejected": -
|
| 214 |
-
"loss": 0.
|
| 215 |
-
"rewards/accuracies": 0.
|
| 216 |
-
"rewards/chosen": -0.
|
| 217 |
-
"rewards/margins":
|
| 218 |
-
"rewards/rejected": -
|
| 219 |
"step": 150
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.33,
|
| 223 |
-
"learning_rate": 9.
|
| 224 |
-
"logits/chosen": 0.
|
| 225 |
-
"logits/rejected": 0.
|
| 226 |
-
"logps/chosen": -
|
| 227 |
-
"logps/rejected": -
|
| 228 |
-
"loss": 0.
|
| 229 |
-
"rewards/accuracies": 0.
|
| 230 |
-
"rewards/chosen": -0.
|
| 231 |
-
"rewards/margins":
|
| 232 |
-
"rewards/rejected": -
|
| 233 |
"step": 160
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"learning_rate": 9.
|
| 238 |
-
"logits/chosen":
|
| 239 |
-
"logits/rejected":
|
| 240 |
-
"logps/chosen": -
|
| 241 |
-
"logps/rejected": -
|
| 242 |
-
"loss": 0.
|
| 243 |
-
"rewards/accuracies": 0.
|
| 244 |
-
"rewards/chosen": -0.
|
| 245 |
-
"rewards/margins":
|
| 246 |
-
"rewards/rejected": -
|
| 247 |
"step": 170
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"learning_rate": 9.
|
| 252 |
-
"logits/chosen": -0.
|
| 253 |
-
"logits/rejected": 0.
|
| 254 |
-
"logps/chosen": -
|
| 255 |
-
"logps/rejected": -
|
| 256 |
-
"loss": 0.
|
| 257 |
-
"rewards/accuracies": 0.
|
| 258 |
-
"rewards/chosen": -0.
|
| 259 |
-
"rewards/margins":
|
| 260 |
-
"rewards/rejected": -
|
| 261 |
"step": 180
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"learning_rate": 9.
|
| 266 |
-
"logits/chosen":
|
| 267 |
-
"logits/rejected": 0.
|
| 268 |
-
"logps/chosen": -
|
| 269 |
-
"logps/rejected": -
|
| 270 |
-
"loss": 0.
|
| 271 |
-
"rewards/accuracies": 0.
|
| 272 |
-
"rewards/chosen": -0.
|
| 273 |
-
"rewards/margins":
|
| 274 |
-
"rewards/rejected": -
|
| 275 |
"step": 190
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 0.
|
| 279 |
-
"learning_rate": 9.
|
| 280 |
-
"logits/chosen":
|
| 281 |
-
"logits/rejected": 0.
|
| 282 |
-
"logps/chosen": -
|
| 283 |
-
"logps/rejected": -
|
| 284 |
-
"loss": 0.
|
| 285 |
-
"rewards/accuracies": 0.
|
| 286 |
-
"rewards/chosen": -
|
| 287 |
-
"rewards/margins":
|
| 288 |
-
"rewards/rejected": -
|
| 289 |
"step": 200
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 0.
|
| 293 |
-
"learning_rate": 9.
|
| 294 |
-
"logits/chosen":
|
| 295 |
-
"logits/rejected": 0.
|
| 296 |
-
"logps/chosen": -
|
| 297 |
-
"logps/rejected": -
|
| 298 |
-
"loss": 0.
|
| 299 |
-
"rewards/accuracies": 0.
|
| 300 |
-
"rewards/chosen": -0.
|
| 301 |
-
"rewards/margins":
|
| 302 |
-
"rewards/rejected": -
|
| 303 |
"step": 210
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"learning_rate": 9.
|
| 308 |
-
"logits/chosen": -0.
|
| 309 |
-
"logits/rejected": -0.
|
| 310 |
-
"logps/chosen": -
|
| 311 |
-
"logps/rejected": -
|
| 312 |
-
"loss": 0.
|
| 313 |
-
"rewards/accuracies": 0.
|
| 314 |
-
"rewards/chosen": -
|
| 315 |
-
"rewards/margins":
|
| 316 |
-
"rewards/rejected": -
|
| 317 |
"step": 220
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 0.
|
| 321 |
-
"learning_rate": 9.
|
| 322 |
-
"logits/chosen":
|
| 323 |
-
"logits/rejected": 0.
|
| 324 |
-
"logps/chosen": -
|
| 325 |
-
"logps/rejected": -
|
| 326 |
-
"loss": 0.
|
| 327 |
-
"rewards/accuracies": 0.
|
| 328 |
-
"rewards/chosen": -
|
| 329 |
-
"rewards/margins":
|
| 330 |
-
"rewards/rejected": -
|
| 331 |
"step": 230
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"learning_rate": 9.
|
| 336 |
-
"logits/chosen":
|
| 337 |
-
"logits/rejected": 0.
|
| 338 |
-
"logps/chosen": -
|
| 339 |
-
"logps/rejected": -
|
| 340 |
-
"loss": 0.
|
| 341 |
"rewards/accuracies": 0.6812499761581421,
|
| 342 |
-
"rewards/chosen": -
|
| 343 |
-
"rewards/margins":
|
| 344 |
-
"rewards/rejected": -
|
| 345 |
"step": 240
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"learning_rate": 9.
|
| 350 |
-
"logits/chosen":
|
| 351 |
-
"logits/rejected": 0.
|
| 352 |
-
"logps/chosen": -
|
| 353 |
-
"logps/rejected": -
|
| 354 |
-
"loss": 0.
|
| 355 |
-
"rewards/accuracies": 0.
|
| 356 |
-
"rewards/chosen": -
|
| 357 |
-
"rewards/margins":
|
| 358 |
-
"rewards/rejected": -
|
| 359 |
"step": 250
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 0.
|
| 363 |
-
"learning_rate": 9.
|
| 364 |
-
"logits/chosen": 0.
|
| 365 |
-
"logits/rejected": 0.
|
| 366 |
-
"logps/chosen": -
|
| 367 |
-
"logps/rejected": -
|
| 368 |
-
"loss": 0.
|
| 369 |
-
"rewards/accuracies": 0.
|
| 370 |
-
"rewards/chosen": -
|
| 371 |
-
"rewards/margins":
|
| 372 |
-
"rewards/rejected": -
|
| 373 |
"step": 260
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 0.
|
| 377 |
-
"learning_rate": 9.
|
| 378 |
-
"logits/chosen":
|
| 379 |
-
"logits/rejected": 0.
|
| 380 |
-
"logps/chosen": -
|
| 381 |
-
"logps/rejected": -
|
| 382 |
-
"loss": 0.
|
| 383 |
"rewards/accuracies": 0.706250011920929,
|
| 384 |
-
"rewards/chosen": -
|
| 385 |
-
"rewards/margins":
|
| 386 |
-
"rewards/rejected": -
|
| 387 |
"step": 270
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 0.
|
| 391 |
-
"learning_rate": 8.
|
| 392 |
-
"logits/chosen":
|
| 393 |
-
"logits/rejected":
|
| 394 |
-
"logps/chosen": -
|
| 395 |
-
"logps/rejected": -
|
| 396 |
-
"loss": 0.
|
| 397 |
-
"rewards/accuracies": 0.
|
| 398 |
-
"rewards/chosen": -
|
| 399 |
-
"rewards/margins":
|
| 400 |
-
"rewards/rejected": -
|
| 401 |
"step": 280
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 0.
|
| 405 |
-
"learning_rate": 8.
|
| 406 |
-
"logits/chosen": 0.
|
| 407 |
-
"logits/rejected": 0.
|
| 408 |
-
"logps/chosen": -
|
| 409 |
-
"logps/rejected": -
|
| 410 |
-
"loss": 0.
|
| 411 |
-
"rewards/accuracies": 0.
|
| 412 |
-
"rewards/chosen": -
|
| 413 |
-
"rewards/margins":
|
| 414 |
-
"rewards/rejected": -
|
| 415 |
"step": 290
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 0.
|
| 419 |
-
"learning_rate": 8.
|
| 420 |
-
"logits/chosen":
|
| 421 |
-
"logits/rejected": 0.
|
| 422 |
-
"logps/chosen": -
|
| 423 |
-
"logps/rejected": -
|
| 424 |
-
"loss": 0.
|
| 425 |
-
"rewards/accuracies": 0.
|
| 426 |
-
"rewards/chosen": -
|
| 427 |
-
"rewards/margins":
|
| 428 |
-
"rewards/rejected": -
|
| 429 |
"step": 300
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 0.
|
| 433 |
-
"learning_rate": 8.
|
| 434 |
-
"logits/chosen": 0.
|
| 435 |
-
"logits/rejected": 0.
|
| 436 |
-
"logps/chosen": -
|
| 437 |
-
"logps/rejected": -
|
| 438 |
-
"loss": 0.
|
| 439 |
-
"rewards/accuracies": 0.
|
| 440 |
-
"rewards/chosen": -
|
| 441 |
-
"rewards/margins":
|
| 442 |
-
"rewards/rejected": -
|
| 443 |
"step": 310
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 0.
|
| 447 |
-
"learning_rate": 8.
|
| 448 |
-
"logits/chosen":
|
| 449 |
-
"logits/rejected": 0.
|
| 450 |
-
"logps/chosen": -
|
| 451 |
-
"logps/rejected": -
|
| 452 |
-
"loss": 0.
|
| 453 |
-
"rewards/accuracies": 0.
|
| 454 |
-
"rewards/chosen": -
|
| 455 |
-
"rewards/margins":
|
| 456 |
-
"rewards/rejected": -
|
| 457 |
"step": 320
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 0.
|
| 461 |
-
"learning_rate": 8.
|
| 462 |
-
"logits/chosen": 0.
|
| 463 |
-
"logits/rejected": 0.
|
| 464 |
-
"logps/chosen": -
|
| 465 |
-
"logps/rejected": -
|
| 466 |
-
"loss": 0.
|
| 467 |
-
"rewards/accuracies": 0.
|
| 468 |
-
"rewards/chosen": -
|
| 469 |
-
"rewards/margins":
|
| 470 |
-
"rewards/rejected": -
|
| 471 |
"step": 330
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 0.
|
| 475 |
-
"learning_rate": 8.
|
| 476 |
-
"logits/chosen":
|
| 477 |
-
"logits/rejected": 0.
|
| 478 |
-
"logps/chosen": -
|
| 479 |
-
"logps/rejected": -
|
| 480 |
-
"loss": 0.
|
| 481 |
-
"rewards/accuracies": 0.
|
| 482 |
-
"rewards/chosen": -
|
| 483 |
-
"rewards/margins":
|
| 484 |
-
"rewards/rejected": -
|
| 485 |
"step": 340
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 0.
|
| 489 |
-
"learning_rate":
|
| 490 |
-
"logits/chosen":
|
| 491 |
-
"logits/rejected":
|
| 492 |
-
"logps/chosen": -
|
| 493 |
-
"logps/rejected": -
|
| 494 |
-
"loss": 0.
|
| 495 |
-
"rewards/accuracies": 0.
|
| 496 |
-
"rewards/chosen": -
|
| 497 |
-
"rewards/margins":
|
| 498 |
-
"rewards/rejected": -
|
| 499 |
"step": 350
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 0.
|
| 503 |
-
"learning_rate": 7.
|
| 504 |
-
"logits/chosen":
|
| 505 |
-
"logits/rejected": 0.
|
| 506 |
-
"logps/chosen": -
|
| 507 |
-
"logps/rejected": -
|
| 508 |
-
"loss": 0.
|
| 509 |
-
"rewards/accuracies": 0.
|
| 510 |
-
"rewards/chosen": -
|
| 511 |
-
"rewards/margins":
|
| 512 |
-
"rewards/rejected": -
|
| 513 |
"step": 360
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 0.
|
| 517 |
-
"learning_rate": 7.
|
| 518 |
-
"logits/chosen":
|
| 519 |
-
"logits/rejected": -0.
|
| 520 |
-
"logps/chosen": -
|
| 521 |
-
"logps/rejected": -
|
| 522 |
-
"loss": 0.
|
| 523 |
-
"rewards/accuracies": 0.
|
| 524 |
-
"rewards/chosen": -
|
| 525 |
-
"rewards/margins":
|
| 526 |
-
"rewards/rejected": -
|
| 527 |
"step": 370
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch": 0.
|
| 531 |
-
"learning_rate": 7.
|
| 532 |
-
"logits/chosen":
|
| 533 |
-
"logits/rejected": -0.
|
| 534 |
-
"logps/chosen": -
|
| 535 |
-
"logps/rejected": -
|
| 536 |
-
"loss": 0.
|
| 537 |
-
"rewards/accuracies": 0.
|
| 538 |
-
"rewards/chosen": -
|
| 539 |
-
"rewards/margins":
|
| 540 |
-
"rewards/rejected": -
|
| 541 |
"step": 380
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch": 0.
|
| 545 |
-
"learning_rate": 7.
|
| 546 |
-
"logits/chosen":
|
| 547 |
-
"logits/rejected": -0.
|
| 548 |
-
"logps/chosen": -
|
| 549 |
-
"logps/rejected": -
|
| 550 |
-
"loss": 0.
|
| 551 |
-
"rewards/accuracies": 0.
|
| 552 |
-
"rewards/chosen": -1.
|
| 553 |
-
"rewards/margins":
|
| 554 |
-
"rewards/rejected": -
|
| 555 |
"step": 390
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 0.
|
| 559 |
-
"learning_rate": 7.
|
| 560 |
-
"logits/chosen":
|
| 561 |
-
"logits/rejected": -0.
|
| 562 |
-
"logps/chosen": -
|
| 563 |
-
"logps/rejected": -
|
| 564 |
-
"loss": 0.
|
| 565 |
-
"rewards/accuracies": 0.
|
| 566 |
-
"rewards/chosen": -
|
| 567 |
-
"rewards/margins":
|
| 568 |
-
"rewards/rejected": -
|
| 569 |
"step": 400
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 0.
|
| 573 |
-
"learning_rate": 7.
|
| 574 |
-
"logits/chosen":
|
| 575 |
-
"logits/rejected": -0.
|
| 576 |
-
"logps/chosen": -
|
| 577 |
-
"logps/rejected": -
|
| 578 |
-
"loss": 0.
|
| 579 |
-
"rewards/accuracies": 0.
|
| 580 |
-
"rewards/chosen": -
|
| 581 |
-
"rewards/margins":
|
| 582 |
-
"rewards/rejected": -
|
| 583 |
"step": 410
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 0.
|
| 587 |
-
"learning_rate":
|
| 588 |
-
"logits/chosen":
|
| 589 |
-
"logits/rejected":
|
| 590 |
-
"logps/chosen": -
|
| 591 |
-
"logps/rejected": -
|
| 592 |
-
"loss": 0.
|
| 593 |
-
"rewards/accuracies": 0.
|
| 594 |
-
"rewards/chosen": -
|
| 595 |
-
"rewards/margins":
|
| 596 |
-
"rewards/rejected": -
|
| 597 |
"step": 420
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 0.
|
| 601 |
-
"learning_rate": 6.
|
| 602 |
-
"logits/chosen":
|
| 603 |
-
"logits/rejected": -0.
|
| 604 |
-
"logps/chosen": -
|
| 605 |
-
"logps/rejected": -
|
| 606 |
-
"loss": 0.
|
| 607 |
-
"rewards/accuracies": 0.
|
| 608 |
-
"rewards/chosen": -
|
| 609 |
-
"rewards/margins":
|
| 610 |
-
"rewards/rejected": -
|
| 611 |
"step": 430
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 0.
|
| 615 |
-
"learning_rate": 6.
|
| 616 |
-
"logits/chosen":
|
| 617 |
-
"logits/rejected": -0.
|
| 618 |
-
"logps/chosen": -
|
| 619 |
-
"logps/rejected": -
|
| 620 |
-
"loss": 0.
|
| 621 |
-
"rewards/accuracies": 0.
|
| 622 |
-
"rewards/chosen": -
|
| 623 |
-
"rewards/margins":
|
| 624 |
-
"rewards/rejected": -
|
| 625 |
"step": 440
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 0.
|
| 629 |
-
"learning_rate": 6.
|
| 630 |
-
"logits/chosen":
|
| 631 |
-
"logits/rejected": -0.
|
| 632 |
-
"logps/chosen": -
|
| 633 |
-
"logps/rejected": -
|
| 634 |
-
"loss": 0.
|
| 635 |
-
"rewards/accuracies": 0.
|
| 636 |
-
"rewards/chosen": -
|
| 637 |
-
"rewards/margins":
|
| 638 |
-
"rewards/rejected": -
|
| 639 |
"step": 450
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 0.
|
| 643 |
-
"learning_rate": 6.
|
| 644 |
-
"logits/chosen":
|
| 645 |
-
"logits/rejected": -0.
|
| 646 |
-
"logps/chosen": -
|
| 647 |
-
"logps/rejected": -
|
| 648 |
-
"loss": 0.
|
| 649 |
-
"rewards/accuracies": 0.
|
| 650 |
-
"rewards/chosen": -
|
| 651 |
-
"rewards/margins":
|
| 652 |
-
"rewards/rejected": -
|
| 653 |
"step": 460
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 0.
|
| 657 |
-
"learning_rate": 6.
|
| 658 |
-
"logits/chosen": -0.
|
| 659 |
-
"logits/rejected": 0.
|
| 660 |
-
"logps/chosen": -
|
| 661 |
-
"logps/rejected": -
|
| 662 |
-
"loss": 0.
|
| 663 |
-
"rewards/accuracies": 0.
|
| 664 |
-
"rewards/chosen": -
|
| 665 |
-
"rewards/margins":
|
| 666 |
-
"rewards/rejected": -
|
| 667 |
"step": 470
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 0
|
| 671 |
-
"learning_rate":
|
| 672 |
-
"logits/chosen":
|
| 673 |
-
"logits/rejected": -0.
|
| 674 |
-
"logps/chosen": -
|
| 675 |
-
"logps/rejected": -
|
| 676 |
-
"loss": 0.
|
| 677 |
-
"rewards/accuracies": 0.
|
| 678 |
-
"rewards/chosen": -
|
| 679 |
-
"rewards/margins":
|
| 680 |
-
"rewards/rejected": -
|
| 681 |
"step": 480
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 1.
|
| 685 |
-
"learning_rate": 5.
|
| 686 |
-
"logits/chosen": -0.
|
| 687 |
-
"logits/rejected": -0.
|
| 688 |
-
"logps/chosen": -
|
| 689 |
-
"logps/rejected": -
|
| 690 |
-
"loss": 0.
|
| 691 |
-
"rewards/accuracies": 0.
|
| 692 |
-
"rewards/chosen": -
|
| 693 |
-
"rewards/margins":
|
| 694 |
-
"rewards/rejected": -
|
| 695 |
"step": 490
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 1.
|
| 699 |
-
"learning_rate": 5.
|
| 700 |
-
"logits/chosen": -0.
|
| 701 |
-
"logits/rejected": 0.
|
| 702 |
-
"logps/chosen": -
|
| 703 |
-
"logps/rejected": -
|
| 704 |
-
"loss": 0.
|
| 705 |
"rewards/accuracies": 0.762499988079071,
|
| 706 |
-
"rewards/chosen": -
|
| 707 |
-
"rewards/margins":
|
| 708 |
-
"rewards/rejected": -
|
| 709 |
"step": 500
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 1.
|
| 713 |
-
"learning_rate": 5.
|
| 714 |
-
"logits/chosen":
|
| 715 |
-
"logits/rejected": -0.
|
| 716 |
-
"logps/chosen": -
|
| 717 |
-
"logps/rejected": -
|
| 718 |
-
"loss": 0.
|
| 719 |
-
"rewards/accuracies": 0.
|
| 720 |
-
"rewards/chosen": -
|
| 721 |
-
"rewards/margins":
|
| 722 |
-
"rewards/rejected": -
|
| 723 |
"step": 510
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch": 1.
|
| 727 |
-
"learning_rate": 5.
|
| 728 |
-
"logits/chosen": -0.
|
| 729 |
-
"logits/rejected": -0.
|
| 730 |
-
"logps/chosen": -
|
| 731 |
-
"logps/rejected": -
|
| 732 |
-
"loss": 0.
|
| 733 |
-
"rewards/accuracies": 0.
|
| 734 |
-
"rewards/chosen": -
|
| 735 |
-
"rewards/margins":
|
| 736 |
-
"rewards/rejected": -
|
| 737 |
"step": 520
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch": 1.
|
| 741 |
-
"learning_rate":
|
| 742 |
-
"logits/chosen":
|
| 743 |
-
"logits/rejected": 0.
|
| 744 |
-
"logps/chosen": -
|
| 745 |
-
"logps/rejected": -
|
| 746 |
-
"loss": 0.
|
| 747 |
-
"rewards/accuracies": 0.
|
| 748 |
-
"rewards/chosen": -
|
| 749 |
-
"rewards/margins":
|
| 750 |
-
"rewards/rejected": -
|
| 751 |
"step": 530
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch": 1.
|
| 755 |
-
"learning_rate": 4.
|
| 756 |
-
"logits/chosen":
|
| 757 |
-
"logits/rejected": -0.
|
| 758 |
-
"logps/chosen": -
|
| 759 |
-
"logps/rejected": -
|
| 760 |
-
"loss": 0.
|
| 761 |
-
"rewards/accuracies": 0.
|
| 762 |
-
"rewards/chosen": -
|
| 763 |
-
"rewards/margins":
|
| 764 |
-
"rewards/rejected": -
|
| 765 |
"step": 540
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch": 1.
|
| 769 |
-
"learning_rate": 4.
|
| 770 |
-
"logits/chosen":
|
| 771 |
-
"logits/rejected": -0.
|
| 772 |
-
"logps/chosen": -
|
| 773 |
-
"logps/rejected": -
|
| 774 |
-
"loss": 0.
|
| 775 |
-
"rewards/accuracies": 0.
|
| 776 |
-
"rewards/chosen": -
|
| 777 |
-
"rewards/margins":
|
| 778 |
-
"rewards/rejected": -
|
| 779 |
"step": 550
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch": 1.
|
| 783 |
-
"learning_rate": 4.
|
| 784 |
-
"logits/chosen": -0.
|
| 785 |
-
"logits/rejected": -0.
|
| 786 |
-
"logps/chosen": -
|
| 787 |
-
"logps/rejected": -
|
| 788 |
-
"loss": 0.
|
| 789 |
"rewards/accuracies": 0.793749988079071,
|
| 790 |
-
"rewards/chosen": -
|
| 791 |
-
"rewards/margins":
|
| 792 |
-
"rewards/rejected": -
|
| 793 |
"step": 560
|
| 794 |
},
|
| 795 |
{
|
| 796 |
-
"epoch": 1.
|
| 797 |
-
"learning_rate": 4.
|
| 798 |
-
"logits/chosen":
|
| 799 |
-
"logits/rejected": -0.
|
| 800 |
-
"logps/chosen": -
|
| 801 |
-
"logps/rejected": -
|
| 802 |
-
"loss": 0.
|
| 803 |
-
"rewards/accuracies": 0.
|
| 804 |
-
"rewards/chosen": -
|
| 805 |
-
"rewards/margins":
|
| 806 |
-
"rewards/rejected": -
|
| 807 |
"step": 570
|
| 808 |
},
|
| 809 |
{
|
| 810 |
-
"epoch": 1.
|
| 811 |
-
"learning_rate":
|
| 812 |
-
"logits/chosen":
|
| 813 |
-
"logits/rejected": 0.
|
| 814 |
-
"logps/chosen": -
|
| 815 |
-
"logps/rejected": -
|
| 816 |
-
"loss": 0.
|
| 817 |
-
"rewards/accuracies": 0.
|
| 818 |
-
"rewards/chosen": -
|
| 819 |
-
"rewards/margins":
|
| 820 |
-
"rewards/rejected": -
|
| 821 |
"step": 580
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch": 1.
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"logits/chosen": 0.
|
| 827 |
-
"logits/rejected": 0.
|
| 828 |
-
"logps/chosen": -
|
| 829 |
-
"logps/rejected": -
|
| 830 |
-
"loss": 0.
|
| 831 |
-
"rewards/accuracies": 0.
|
| 832 |
-
"rewards/chosen": -
|
| 833 |
-
"rewards/margins":
|
| 834 |
-
"rewards/rejected": -
|
| 835 |
"step": 590
|
| 836 |
},
|
| 837 |
{
|
| 838 |
-
"epoch": 1.
|
| 839 |
-
"learning_rate": 3.
|
| 840 |
-
"logits/chosen":
|
| 841 |
-
"logits/rejected": 0.
|
| 842 |
-
"logps/chosen": -
|
| 843 |
-
"logps/rejected": -
|
| 844 |
-
"loss": 0.
|
| 845 |
-
"rewards/accuracies": 0.
|
| 846 |
-
"rewards/chosen": -
|
| 847 |
-
"rewards/margins":
|
| 848 |
-
"rewards/rejected": -
|
| 849 |
"step": 600
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch": 1.
|
| 853 |
-
"learning_rate": 3.
|
| 854 |
-
"logits/chosen": 0.
|
| 855 |
-
"logits/rejected": 0.
|
| 856 |
-
"logps/chosen": -
|
| 857 |
-
"logps/rejected": -
|
| 858 |
-
"loss": 0.
|
| 859 |
-
"rewards/accuracies": 0.
|
| 860 |
-
"rewards/chosen": -
|
| 861 |
-
"rewards/margins":
|
| 862 |
-
"rewards/rejected": -
|
| 863 |
"step": 610
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch": 1.
|
| 867 |
-
"learning_rate": 3.
|
| 868 |
-
"logits/chosen": 0.
|
| 869 |
-
"logits/rejected": 0.
|
| 870 |
-
"logps/chosen": -
|
| 871 |
-
"logps/rejected": -
|
| 872 |
-
"loss": 0.
|
| 873 |
-
"rewards/accuracies": 0.
|
| 874 |
-
"rewards/chosen": -
|
| 875 |
-
"rewards/margins":
|
| 876 |
-
"rewards/rejected": -
|
| 877 |
"step": 620
|
| 878 |
},
|
| 879 |
{
|
| 880 |
-
"epoch": 1.
|
| 881 |
-
"learning_rate": 3.
|
| 882 |
-
"logits/chosen": 0.
|
| 883 |
-
"logits/rejected": 0.
|
| 884 |
-
"logps/chosen": -
|
| 885 |
-
"logps/rejected": -
|
| 886 |
-
"loss": 0.
|
| 887 |
-
"rewards/accuracies": 0.
|
| 888 |
-
"rewards/chosen": -
|
| 889 |
-
"rewards/margins":
|
| 890 |
-
"rewards/rejected": -
|
| 891 |
"step": 630
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"epoch": 1.
|
| 895 |
-
"learning_rate":
|
| 896 |
-
"logits/chosen": 0.
|
| 897 |
-
"logits/rejected": 0.
|
| 898 |
-
"logps/chosen": -
|
| 899 |
-
"logps/rejected": -
|
| 900 |
-
"loss": 0.
|
| 901 |
-
"rewards/accuracies": 0.
|
| 902 |
-
"rewards/chosen": -
|
| 903 |
-
"rewards/margins":
|
| 904 |
-
"rewards/rejected": -
|
| 905 |
"step": 640
|
| 906 |
},
|
| 907 |
{
|
| 908 |
-
"epoch": 1.
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"logits/chosen": 0.
|
| 911 |
-
"logits/rejected": 0.
|
| 912 |
-
"logps/chosen": -
|
| 913 |
-
"logps/rejected": -
|
| 914 |
-
"loss": 0.
|
| 915 |
-
"rewards/accuracies": 0.
|
| 916 |
-
"rewards/chosen": -2.
|
| 917 |
-
"rewards/margins":
|
| 918 |
-
"rewards/rejected": -
|
| 919 |
"step": 650
|
| 920 |
},
|
| 921 |
{
|
| 922 |
-
"epoch": 1.
|
| 923 |
-
"learning_rate": 2.
|
| 924 |
-
"logits/chosen": 0.
|
| 925 |
-
"logits/rejected": 0.
|
| 926 |
-
"logps/chosen": -
|
| 927 |
-
"logps/rejected": -
|
| 928 |
-
"loss": 0.
|
| 929 |
-
"rewards/accuracies": 0.
|
| 930 |
-
"rewards/chosen": -
|
| 931 |
-
"rewards/margins":
|
| 932 |
-
"rewards/rejected": -
|
| 933 |
"step": 660
|
| 934 |
},
|
| 935 |
{
|
| 936 |
-
"epoch": 1.
|
| 937 |
-
"learning_rate": 2.
|
| 938 |
-
"logits/chosen": 0.
|
| 939 |
-
"logits/rejected": 0.
|
| 940 |
-
"logps/chosen": -
|
| 941 |
-
"logps/rejected": -
|
| 942 |
-
"loss": 0.
|
| 943 |
"rewards/accuracies": 0.768750011920929,
|
| 944 |
-
"rewards/chosen": -
|
| 945 |
-
"rewards/margins":
|
| 946 |
-
"rewards/rejected": -
|
| 947 |
"step": 670
|
| 948 |
},
|
| 949 |
{
|
| 950 |
-
"epoch": 1.
|
| 951 |
-
"learning_rate": 2.
|
| 952 |
-
"logits/chosen": 0.
|
| 953 |
-
"logits/rejected": 0.
|
| 954 |
-
"logps/chosen": -
|
| 955 |
-
"logps/rejected": -
|
| 956 |
-
"loss": 0.
|
| 957 |
-
"rewards/accuracies": 0.
|
| 958 |
-
"rewards/chosen": -
|
| 959 |
-
"rewards/margins":
|
| 960 |
-
"rewards/rejected": -
|
| 961 |
"step": 680
|
| 962 |
},
|
| 963 |
{
|
| 964 |
-
"epoch": 1.
|
| 965 |
-
"learning_rate": 2.
|
| 966 |
-
"logits/chosen": 0.
|
| 967 |
-
"logits/rejected": 0.
|
| 968 |
-
"logps/chosen": -
|
| 969 |
-
"logps/rejected": -
|
| 970 |
-
"loss": 0.
|
| 971 |
-
"rewards/accuracies": 0.
|
| 972 |
-
"rewards/chosen": -
|
| 973 |
-
"rewards/margins":
|
| 974 |
-
"rewards/rejected": -
|
| 975 |
"step": 690
|
| 976 |
},
|
| 977 |
{
|
| 978 |
-
"epoch": 1.
|
| 979 |
-
"learning_rate": 2.
|
| 980 |
-
"logits/chosen": 0.
|
| 981 |
-
"logits/rejected": 0.
|
| 982 |
-
"logps/chosen": -
|
| 983 |
-
"logps/rejected": -
|
| 984 |
-
"loss": 0.
|
| 985 |
-
"rewards/accuracies": 0.
|
| 986 |
-
"rewards/chosen": -
|
| 987 |
-
"rewards/margins":
|
| 988 |
-
"rewards/rejected": -
|
| 989 |
"step": 700
|
| 990 |
},
|
| 991 |
{
|
| 992 |
-
"epoch": 1.
|
| 993 |
-
"learning_rate":
|
| 994 |
-
"logits/chosen": 0.
|
| 995 |
-
"logits/rejected": 0.
|
| 996 |
-
"logps/chosen": -
|
| 997 |
-
"logps/rejected": -
|
| 998 |
-
"loss": 0.
|
| 999 |
-
"rewards/accuracies": 0.
|
| 1000 |
-
"rewards/chosen": -
|
| 1001 |
-
"rewards/margins":
|
| 1002 |
-
"rewards/rejected": -
|
| 1003 |
"step": 710
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
-
"epoch": 1.
|
| 1007 |
-
"learning_rate": 1.
|
| 1008 |
-
"logits/chosen": 0.
|
| 1009 |
-
"logits/rejected": 0.
|
| 1010 |
-
"logps/chosen": -
|
| 1011 |
-
"logps/rejected": -
|
| 1012 |
-
"loss": 0.
|
| 1013 |
-
"rewards/accuracies": 0.
|
| 1014 |
-
"rewards/chosen": -
|
| 1015 |
-
"rewards/margins":
|
| 1016 |
-
"rewards/rejected": -
|
| 1017 |
"step": 720
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
-
"epoch": 1.
|
| 1021 |
-
"learning_rate": 1.
|
| 1022 |
-
"logits/chosen": 0.
|
| 1023 |
-
"logits/rejected": 0.
|
| 1024 |
-
"logps/chosen": -
|
| 1025 |
-
"logps/rejected": -
|
| 1026 |
-
"loss": 0.
|
| 1027 |
-
"rewards/accuracies": 0.
|
| 1028 |
-
"rewards/chosen": -
|
| 1029 |
-
"rewards/margins":
|
| 1030 |
-
"rewards/rejected": -
|
| 1031 |
"step": 730
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
-
"epoch": 1.
|
| 1035 |
-
"learning_rate": 1.
|
| 1036 |
-
"logits/chosen": 0.
|
| 1037 |
-
"logits/rejected": 0.
|
| 1038 |
-
"logps/chosen": -
|
| 1039 |
-
"logps/rejected": -
|
| 1040 |
-
"loss": 0.
|
| 1041 |
-
"rewards/accuracies": 0.
|
| 1042 |
-
"rewards/chosen": -
|
| 1043 |
-
"rewards/margins":
|
| 1044 |
-
"rewards/rejected": -
|
| 1045 |
"step": 740
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
-
"epoch": 1.
|
| 1049 |
-
"learning_rate": 1.
|
| 1050 |
-
"logits/chosen": 0.
|
| 1051 |
-
"logits/rejected": 0.
|
| 1052 |
-
"logps/chosen": -
|
| 1053 |
-
"logps/rejected": -
|
| 1054 |
-
"loss": 0.
|
| 1055 |
-
"rewards/accuracies": 0.
|
| 1056 |
-
"rewards/chosen": -
|
| 1057 |
-
"rewards/margins":
|
| 1058 |
-
"rewards/rejected": -
|
| 1059 |
"step": 750
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
-
"epoch": 1.
|
| 1063 |
-
"learning_rate": 1.
|
| 1064 |
-
"logits/chosen": 0.
|
| 1065 |
-
"logits/rejected": 0.
|
| 1066 |
-
"logps/chosen": -
|
| 1067 |
-
"logps/rejected": -
|
| 1068 |
-
"loss": 0.
|
| 1069 |
-
"rewards/accuracies": 0.
|
| 1070 |
-
"rewards/chosen": -
|
| 1071 |
-
"rewards/margins":
|
| 1072 |
-
"rewards/rejected": -
|
| 1073 |
"step": 760
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
-
"epoch": 1.
|
| 1077 |
-
"learning_rate": 1.
|
| 1078 |
-
"logits/chosen": 0.
|
| 1079 |
-
"logits/rejected": 0.
|
| 1080 |
-
"logps/chosen": -
|
| 1081 |
-
"logps/rejected": -
|
| 1082 |
-
"loss": 0.
|
| 1083 |
-
"rewards/accuracies": 0.
|
| 1084 |
-
"rewards/chosen": -
|
| 1085 |
-
"rewards/margins":
|
| 1086 |
-
"rewards/rejected": -
|
| 1087 |
"step": 770
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
-
"epoch": 1.
|
| 1091 |
-
"learning_rate":
|
| 1092 |
-
"logits/chosen": 0.
|
| 1093 |
-
"logits/rejected": 0.
|
| 1094 |
-
"logps/chosen": -
|
| 1095 |
-
"logps/rejected": -
|
| 1096 |
-
"loss": 0.
|
| 1097 |
-
"rewards/accuracies": 0.
|
| 1098 |
-
"rewards/chosen": -
|
| 1099 |
-
"rewards/margins":
|
| 1100 |
-
"rewards/rejected": -
|
| 1101 |
"step": 780
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
-
"epoch": 1.
|
| 1105 |
-
"learning_rate":
|
| 1106 |
-
"logits/chosen": 0.
|
| 1107 |
-
"logits/rejected": 0.
|
| 1108 |
-
"logps/chosen": -
|
| 1109 |
-
"logps/rejected": -
|
| 1110 |
-
"loss": 0.
|
| 1111 |
-
"rewards/accuracies": 0.
|
| 1112 |
-
"rewards/chosen": -
|
| 1113 |
-
"rewards/margins":
|
| 1114 |
-
"rewards/rejected": -
|
| 1115 |
"step": 790
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
-
"epoch": 1.
|
| 1119 |
-
"learning_rate":
|
| 1120 |
-
"logits/chosen": 0.
|
| 1121 |
-
"logits/rejected": 0.
|
| 1122 |
-
"logps/chosen": -
|
| 1123 |
-
"logps/rejected": -
|
| 1124 |
-
"loss": 0.
|
| 1125 |
-
"rewards/accuracies": 0.
|
| 1126 |
-
"rewards/chosen": -
|
| 1127 |
-
"rewards/margins":
|
| 1128 |
-
"rewards/rejected": -
|
| 1129 |
"step": 800
|
| 1130 |
},
|
| 1131 |
-
{
|
| 1132 |
-
"epoch": 1.66,
|
| 1133 |
-
"learning_rate": 8.563690047752148e-08,
|
| 1134 |
-
"logits/chosen": 0.20595593750476837,
|
| 1135 |
-
"logits/rejected": 0.3465935289859772,
|
| 1136 |
-
"logps/chosen": -603.9895629882812,
|
| 1137 |
-
"logps/rejected": -659.3458862304688,
|
| 1138 |
-
"loss": 0.0281,
|
| 1139 |
-
"rewards/accuracies": 0.800000011920929,
|
| 1140 |
-
"rewards/chosen": -3.0521774291992188,
|
| 1141 |
-
"rewards/margins": 1.057278037071228,
|
| 1142 |
-
"rewards/rejected": -4.109455585479736,
|
| 1143 |
-
"step": 810
|
| 1144 |
-
},
|
| 1145 |
-
{
|
| 1146 |
-
"epoch": 1.68,
|
| 1147 |
-
"learning_rate": 7.589170151660656e-08,
|
| 1148 |
-
"logits/chosen": 0.2923261523246765,
|
| 1149 |
-
"logits/rejected": 0.49081581830978394,
|
| 1150 |
-
"logps/chosen": -580.5206909179688,
|
| 1151 |
-
"logps/rejected": -572.1239013671875,
|
| 1152 |
-
"loss": 0.03,
|
| 1153 |
-
"rewards/accuracies": 0.71875,
|
| 1154 |
-
"rewards/chosen": -2.961920976638794,
|
| 1155 |
-
"rewards/margins": 0.8716068267822266,
|
| 1156 |
-
"rewards/rejected": -3.8335273265838623,
|
| 1157 |
-
"step": 820
|
| 1158 |
-
},
|
| 1159 |
{
|
| 1160 |
"epoch": 1.7,
|
| 1161 |
-
"learning_rate": 6.
|
| 1162 |
-
"logits/chosen": 0.
|
| 1163 |
-
"logits/rejected": 0.
|
| 1164 |
-
"logps/chosen": -
|
| 1165 |
-
"logps/rejected": -
|
| 1166 |
-
"loss": 0.
|
| 1167 |
-
"rewards/accuracies": 0.
|
| 1168 |
-
"rewards/chosen": -
|
| 1169 |
-
"rewards/margins":
|
| 1170 |
-
"rewards/rejected": -
|
| 1171 |
-
"step":
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 1.72,
|
| 1175 |
-
"learning_rate": 5.
|
| 1176 |
-
"logits/chosen": 0.
|
| 1177 |
-
"logits/rejected": 0.
|
| 1178 |
-
"logps/chosen": -
|
| 1179 |
-
"logps/rejected": -
|
| 1180 |
-
"loss": 0.
|
| 1181 |
-
"rewards/accuracies": 0.
|
| 1182 |
-
"rewards/chosen": -
|
| 1183 |
-
"rewards/margins":
|
| 1184 |
-
"rewards/rejected": -
|
| 1185 |
-
"step":
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 1.74,
|
| 1189 |
-
"learning_rate":
|
| 1190 |
-
"logits/chosen": 0.
|
| 1191 |
-
"logits/rejected": 0.
|
| 1192 |
-
"logps/chosen": -
|
| 1193 |
-
"logps/rejected": -
|
| 1194 |
-
"loss": 0.
|
| 1195 |
-
"rewards/accuracies": 0.
|
| 1196 |
-
"rewards/chosen": -
|
| 1197 |
-
"rewards/margins":
|
| 1198 |
-
"rewards/rejected": -
|
| 1199 |
-
"step":
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 1.76,
|
| 1203 |
-
"learning_rate": 4.
|
| 1204 |
-
"logits/chosen": 0.
|
| 1205 |
-
"logits/rejected": 0.
|
| 1206 |
-
"logps/chosen": -
|
| 1207 |
-
"logps/rejected": -
|
| 1208 |
-
"loss": 0.
|
| 1209 |
-
"rewards/accuracies": 0.
|
| 1210 |
-
"rewards/chosen": -
|
| 1211 |
-
"rewards/margins":
|
| 1212 |
-
"rewards/rejected": -
|
| 1213 |
-
"step":
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 1.78,
|
| 1217 |
-
"learning_rate": 3.
|
| 1218 |
-
"logits/chosen": 0.
|
| 1219 |
-
"logits/rejected": 0.
|
| 1220 |
-
"logps/chosen": -
|
| 1221 |
-
"logps/rejected": -
|
| 1222 |
-
"loss": 0.
|
| 1223 |
-
"rewards/accuracies": 0.
|
| 1224 |
-
"rewards/chosen": -
|
| 1225 |
-
"rewards/margins":
|
| 1226 |
-
"rewards/rejected": -
|
| 1227 |
-
"step":
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 1.8,
|
| 1231 |
-
"learning_rate": 2.
|
| 1232 |
-
"logits/chosen": 0.
|
| 1233 |
-
"logits/rejected": 0.
|
| 1234 |
-
"logps/chosen": -
|
| 1235 |
-
"logps/rejected": -
|
| 1236 |
-
"loss": 0.
|
| 1237 |
-
"rewards/accuracies": 0.
|
| 1238 |
-
"rewards/chosen": -
|
| 1239 |
-
"rewards/margins":
|
| 1240 |
-
"rewards/rejected": -
|
| 1241 |
-
"step":
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 1.82,
|
| 1245 |
-
"learning_rate": 2.
|
| 1246 |
-
"logits/chosen": 0.
|
| 1247 |
-
"logits/rejected": 0.
|
| 1248 |
-
"logps/chosen": -
|
| 1249 |
-
"logps/rejected": -
|
| 1250 |
-
"loss": 0.
|
| 1251 |
-
"rewards/accuracies": 0.
|
| 1252 |
-
"rewards/chosen": -
|
| 1253 |
-
"rewards/margins":
|
| 1254 |
-
"rewards/rejected": -
|
| 1255 |
-
"step":
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
"epoch": 1.84,
|
| 1259 |
-
"learning_rate": 1.
|
| 1260 |
-
"logits/chosen": 0.
|
| 1261 |
-
"logits/rejected": 0.
|
| 1262 |
-
"logps/chosen": -
|
| 1263 |
-
"logps/rejected": -
|
| 1264 |
-
"loss": 0.
|
| 1265 |
"rewards/accuracies": 0.731249988079071,
|
| 1266 |
-
"rewards/chosen": -
|
| 1267 |
-
"rewards/margins":
|
| 1268 |
-
"rewards/rejected": -
|
| 1269 |
-
"step":
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 1.86,
|
| 1273 |
-
"learning_rate": 1.
|
| 1274 |
-
"logits/chosen": 0.
|
| 1275 |
-
"logits/rejected": 0.
|
| 1276 |
-
"logps/chosen": -
|
| 1277 |
-
"logps/rejected": -
|
| 1278 |
-
"loss": 0.
|
| 1279 |
-
"rewards/accuracies": 0.
|
| 1280 |
-
"rewards/chosen": -
|
| 1281 |
-
"rewards/margins":
|
| 1282 |
-
"rewards/rejected": -
|
| 1283 |
-
"step":
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
"epoch": 1.88,
|
| 1287 |
-
"learning_rate":
|
| 1288 |
-
"logits/chosen": 0.
|
| 1289 |
-
"logits/rejected": 0.
|
| 1290 |
-
"logps/chosen": -
|
| 1291 |
-
"logps/rejected": -
|
| 1292 |
-
"loss": 0.
|
| 1293 |
-
"rewards/accuracies": 0.
|
| 1294 |
-
"rewards/chosen": -2.
|
| 1295 |
-
"rewards/margins":
|
| 1296 |
-
"rewards/rejected": -
|
| 1297 |
-
"step":
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
"epoch": 1.91,
|
| 1301 |
-
"learning_rate": 6.
|
| 1302 |
-
"logits/chosen": 0.
|
| 1303 |
-
"logits/rejected": 0.
|
| 1304 |
-
"logps/chosen": -
|
| 1305 |
-
"logps/rejected": -
|
| 1306 |
-
"loss": 0.
|
| 1307 |
-
"rewards/accuracies": 0.
|
| 1308 |
-
"rewards/chosen": -
|
| 1309 |
-
"rewards/margins":
|
| 1310 |
-
"rewards/rejected": -
|
| 1311 |
-
"step":
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
"epoch": 1.93,
|
| 1315 |
-
"learning_rate":
|
| 1316 |
-
"logits/chosen": 0.
|
| 1317 |
-
"logits/rejected": 0.
|
| 1318 |
-
"logps/chosen": -
|
| 1319 |
-
"logps/rejected": -
|
| 1320 |
-
"loss": 0.
|
| 1321 |
-
"rewards/accuracies": 0.
|
| 1322 |
-
"rewards/chosen": -
|
| 1323 |
-
"rewards/margins":
|
| 1324 |
-
"rewards/rejected": -
|
| 1325 |
-
"step":
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
"epoch": 1.95,
|
| 1329 |
-
"learning_rate":
|
| 1330 |
-
"logits/chosen": 0.
|
| 1331 |
-
"logits/rejected": 0.
|
| 1332 |
-
"logps/chosen": -
|
| 1333 |
-
"logps/rejected": -
|
| 1334 |
-
"loss": 0.
|
| 1335 |
-
"rewards/accuracies": 0.
|
| 1336 |
-
"rewards/chosen": -
|
| 1337 |
-
"rewards/margins":
|
| 1338 |
-
"rewards/rejected": -
|
| 1339 |
-
"step":
|
| 1340 |
},
|
| 1341 |
{
|
| 1342 |
"epoch": 1.97,
|
| 1343 |
-
"learning_rate":
|
| 1344 |
-
"logits/chosen": 0.
|
| 1345 |
-
"logits/rejected": 0.
|
| 1346 |
-
"logps/chosen": -
|
| 1347 |
-
"logps/rejected": -
|
| 1348 |
-
"loss": 0.
|
| 1349 |
-
"rewards/accuracies": 0.
|
| 1350 |
-
"rewards/chosen": -
|
| 1351 |
-
"rewards/margins":
|
| 1352 |
-
"rewards/rejected": -
|
| 1353 |
-
"step":
|
| 1354 |
},
|
| 1355 |
{
|
| 1356 |
"epoch": 1.99,
|
| 1357 |
-
"learning_rate":
|
| 1358 |
-
"logits/chosen": 0.
|
| 1359 |
-
"logits/rejected": 0.
|
| 1360 |
-
"logps/chosen": -
|
| 1361 |
-
"logps/rejected": -
|
| 1362 |
-
"loss": 0.
|
| 1363 |
-
"rewards/accuracies": 0.
|
| 1364 |
-
"rewards/chosen": -
|
| 1365 |
-
"rewards/margins":
|
| 1366 |
-
"rewards/rejected": -
|
| 1367 |
-
"step":
|
| 1368 |
},
|
| 1369 |
{
|
| 1370 |
"epoch": 2.0,
|
| 1371 |
-
"step":
|
| 1372 |
"total_flos": 0.0,
|
| 1373 |
-
"train_loss": 0.
|
| 1374 |
-
"train_runtime":
|
| 1375 |
-
"train_samples_per_second":
|
| 1376 |
-
"train_steps_per_second": 0.
|
| 1377 |
}
|
| 1378 |
],
|
| 1379 |
"logging_steps": 10,
|
| 1380 |
-
"max_steps":
|
| 1381 |
"num_train_epochs": 2,
|
| 1382 |
"save_steps": 10000,
|
| 1383 |
"total_flos": 0.0,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9973828840617638,
|
| 5 |
"eval_steps": 10000,
|
| 6 |
+
"global_step": 954,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 1.0416666666666667e-07,
|
| 14 |
+
"logits/chosen": 0.17733711004257202,
|
| 15 |
+
"logits/rejected": 0.2543194591999054,
|
| 16 |
+
"logps/chosen": -354.2892150878906,
|
| 17 |
+
"logps/rejected": -305.198974609375,
|
| 18 |
+
"loss": 0.5,
|
| 19 |
+
"rewards/accuracies": 0.40625,
|
| 20 |
+
"rewards/chosen": 0.001110748155042529,
|
| 21 |
+
"rewards/margins": 0.0009211620199494064,
|
| 22 |
+
"rewards/rejected": 0.0001895862224046141,
|
| 23 |
"step": 10
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.04,
|
| 27 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 28 |
+
"logits/chosen": 0.0703902393579483,
|
| 29 |
+
"logits/rejected": 0.19805452227592468,
|
| 30 |
+
"logps/chosen": -316.7099609375,
|
| 31 |
+
"logps/rejected": -276.1195373535156,
|
| 32 |
+
"loss": 0.4999,
|
| 33 |
+
"rewards/accuracies": 0.543749988079071,
|
| 34 |
+
"rewards/chosen": 0.0002531521604396403,
|
| 35 |
+
"rewards/margins": 0.001093686674721539,
|
| 36 |
+
"rewards/rejected": -0.0008405345724895597,
|
| 37 |
"step": 20
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.06,
|
| 41 |
+
"learning_rate": 3.1249999999999997e-07,
|
| 42 |
+
"logits/chosen": 0.1768152415752411,
|
| 43 |
+
"logits/rejected": 0.24860361218452454,
|
| 44 |
+
"logps/chosen": -294.9255676269531,
|
| 45 |
+
"logps/rejected": -298.43670654296875,
|
| 46 |
+
"loss": 0.4995,
|
| 47 |
+
"rewards/accuracies": 0.59375,
|
| 48 |
+
"rewards/chosen": 0.0004133238398935646,
|
| 49 |
+
"rewards/margins": 0.0018008403712883592,
|
| 50 |
+
"rewards/rejected": -0.001387516618706286,
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.08,
|
| 55 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 56 |
+
"logits/chosen": 0.09639827907085419,
|
| 57 |
+
"logits/rejected": 0.2179565727710724,
|
| 58 |
+
"logps/chosen": -346.4007263183594,
|
| 59 |
+
"logps/rejected": -320.07464599609375,
|
| 60 |
+
"loss": 0.4982,
|
| 61 |
+
"rewards/accuracies": 0.643750011920929,
|
| 62 |
+
"rewards/chosen": 0.011715102009475231,
|
| 63 |
+
"rewards/margins": 0.0077905962243676186,
|
| 64 |
+
"rewards/rejected": 0.003924505319446325,
|
| 65 |
"step": 40
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.1,
|
| 69 |
+
"learning_rate": 5.208333333333334e-07,
|
| 70 |
+
"logits/chosen": 0.14820781350135803,
|
| 71 |
+
"logits/rejected": 0.23618468642234802,
|
| 72 |
+
"logps/chosen": -308.26483154296875,
|
| 73 |
+
"logps/rejected": -283.5643310546875,
|
| 74 |
+
"loss": 0.496,
|
| 75 |
+
"rewards/accuracies": 0.6812499761581421,
|
| 76 |
+
"rewards/chosen": 0.022877948358654976,
|
| 77 |
+
"rewards/margins": 0.021698923781514168,
|
| 78 |
+
"rewards/rejected": 0.0011790238786488771,
|
| 79 |
"step": 50
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.13,
|
| 83 |
+
"learning_rate": 6.249999999999999e-07,
|
| 84 |
+
"logits/chosen": 0.12726549804210663,
|
| 85 |
+
"logits/rejected": 0.2663690447807312,
|
| 86 |
+
"logps/chosen": -289.76361083984375,
|
| 87 |
+
"logps/rejected": -275.6463623046875,
|
| 88 |
+
"loss": 0.4922,
|
| 89 |
+
"rewards/accuracies": 0.6875,
|
| 90 |
+
"rewards/chosen": 0.031151825562119484,
|
| 91 |
+
"rewards/margins": 0.033161893486976624,
|
| 92 |
+
"rewards/rejected": -0.0020100646652281284,
|
| 93 |
"step": 60
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.15,
|
| 97 |
+
"learning_rate": 7.291666666666666e-07,
|
| 98 |
+
"logits/chosen": 0.1551138460636139,
|
| 99 |
+
"logits/rejected": 0.2375030517578125,
|
| 100 |
+
"logps/chosen": -323.76788330078125,
|
| 101 |
+
"logps/rejected": -318.1293029785156,
|
| 102 |
+
"loss": 0.4879,
|
| 103 |
+
"rewards/accuracies": 0.59375,
|
| 104 |
+
"rewards/chosen": 0.059557922184467316,
|
| 105 |
+
"rewards/margins": 0.05137287825345993,
|
| 106 |
+
"rewards/rejected": 0.008185049518942833,
|
| 107 |
"step": 70
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.17,
|
| 111 |
+
"learning_rate": 8.333333333333333e-07,
|
| 112 |
+
"logits/chosen": 0.22313980758190155,
|
| 113 |
+
"logits/rejected": 0.31770533323287964,
|
| 114 |
+
"logps/chosen": -312.06854248046875,
|
| 115 |
+
"logps/rejected": -291.3954162597656,
|
| 116 |
+
"loss": 0.4823,
|
| 117 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 118 |
+
"rewards/chosen": 0.06595131754875183,
|
| 119 |
+
"rewards/margins": 0.10923006385564804,
|
| 120 |
+
"rewards/rejected": -0.04327874630689621,
|
| 121 |
"step": 80
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.19,
|
| 125 |
+
"learning_rate": 9.374999999999999e-07,
|
| 126 |
+
"logits/chosen": 0.111533522605896,
|
| 127 |
+
"logits/rejected": 0.25437992811203003,
|
| 128 |
+
"logps/chosen": -331.6070556640625,
|
| 129 |
+
"logps/rejected": -276.5615234375,
|
| 130 |
+
"loss": 0.4763,
|
| 131 |
+
"rewards/accuracies": 0.6812499761581421,
|
| 132 |
+
"rewards/chosen": 0.05334480479359627,
|
| 133 |
+
"rewards/margins": 0.14821472764015198,
|
| 134 |
+
"rewards/rejected": -0.09486991912126541,
|
| 135 |
"step": 90
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.21,
|
| 139 |
+
"learning_rate": 9.999463737538052e-07,
|
| 140 |
+
"logits/chosen": 0.19283509254455566,
|
| 141 |
+
"logits/rejected": 0.36483508348464966,
|
| 142 |
+
"logps/chosen": -334.4452819824219,
|
| 143 |
+
"logps/rejected": -312.9904479980469,
|
| 144 |
+
"loss": 0.4681,
|
| 145 |
"rewards/accuracies": 0.643750011920929,
|
| 146 |
+
"rewards/chosen": 0.0011617511045187712,
|
| 147 |
+
"rewards/margins": 0.20724515616893768,
|
| 148 |
+
"rewards/rejected": -0.20608338713645935,
|
| 149 |
"step": 100
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.23,
|
| 153 |
+
"learning_rate": 9.993432105822034e-07,
|
| 154 |
+
"logits/chosen": 0.21006178855895996,
|
| 155 |
+
"logits/rejected": 0.2403305023908615,
|
| 156 |
+
"logps/chosen": -329.19293212890625,
|
| 157 |
+
"logps/rejected": -345.82330322265625,
|
| 158 |
+
"loss": 0.457,
|
| 159 |
+
"rewards/accuracies": 0.675000011920929,
|
| 160 |
+
"rewards/chosen": -0.16573527455329895,
|
| 161 |
+
"rewards/margins": 0.2979966104030609,
|
| 162 |
+
"rewards/rejected": -0.4637318551540375,
|
| 163 |
"step": 110
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.25,
|
| 167 |
+
"learning_rate": 9.980706626858607e-07,
|
| 168 |
+
"logits/chosen": 0.15046869218349457,
|
| 169 |
+
"logits/rejected": 0.21053044497966766,
|
| 170 |
+
"logps/chosen": -360.9029846191406,
|
| 171 |
+
"logps/rejected": -405.44732666015625,
|
| 172 |
+
"loss": 0.46,
|
| 173 |
+
"rewards/accuracies": 0.637499988079071,
|
| 174 |
+
"rewards/chosen": -0.40349340438842773,
|
| 175 |
+
"rewards/margins": 0.43901604413986206,
|
| 176 |
+
"rewards/rejected": -0.8425094485282898,
|
| 177 |
"step": 120
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.27,
|
| 181 |
+
"learning_rate": 9.961304359538434e-07,
|
| 182 |
+
"logits/chosen": 0.13695412874221802,
|
| 183 |
+
"logits/rejected": 0.2686176002025604,
|
| 184 |
+
"logps/chosen": -424.5625915527344,
|
| 185 |
+
"logps/rejected": -432.45745849609375,
|
| 186 |
+
"loss": 0.4508,
|
| 187 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 188 |
+
"rewards/chosen": -0.8606651425361633,
|
| 189 |
+
"rewards/margins": 0.6154058575630188,
|
| 190 |
+
"rewards/rejected": -1.4760708808898926,
|
| 191 |
"step": 130
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.29,
|
| 195 |
+
"learning_rate": 9.935251313189563e-07,
|
| 196 |
+
"logits/chosen": 0.08335243165493011,
|
| 197 |
+
"logits/rejected": 0.23106291890144348,
|
| 198 |
+
"logps/chosen": -490.5091247558594,
|
| 199 |
+
"logps/rejected": -535.1593017578125,
|
| 200 |
+
"loss": 0.4468,
|
| 201 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 202 |
+
"rewards/chosen": -1.5846575498580933,
|
| 203 |
+
"rewards/margins": 1.0678659677505493,
|
| 204 |
+
"rewards/rejected": -2.6525235176086426,
|
| 205 |
"step": 140
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.31,
|
| 209 |
+
"learning_rate": 9.902582412711118e-07,
|
| 210 |
+
"logits/chosen": 0.16554813086986542,
|
| 211 |
+
"logits/rejected": 0.20584425330162048,
|
| 212 |
+
"logps/chosen": -427.4266052246094,
|
| 213 |
+
"logps/rejected": -530.4254150390625,
|
| 214 |
+
"loss": 0.445,
|
| 215 |
+
"rewards/accuracies": 0.71875,
|
| 216 |
+
"rewards/chosen": -0.9212606549263,
|
| 217 |
+
"rewards/margins": 1.2334972620010376,
|
| 218 |
+
"rewards/rejected": -2.1547577381134033,
|
| 219 |
"step": 150
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.33,
|
| 223 |
+
"learning_rate": 9.86334145175542e-07,
|
| 224 |
+
"logits/chosen": 0.08956819772720337,
|
| 225 |
+
"logits/rejected": 0.10614663362503052,
|
| 226 |
+
"logps/chosen": -355.2901916503906,
|
| 227 |
+
"logps/rejected": -487.94354248046875,
|
| 228 |
+
"loss": 0.4399,
|
| 229 |
+
"rewards/accuracies": 0.731249988079071,
|
| 230 |
+
"rewards/chosen": -0.5322867631912231,
|
| 231 |
+
"rewards/margins": 1.5401604175567627,
|
| 232 |
+
"rewards/rejected": -2.0724472999572754,
|
| 233 |
"step": 160
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.36,
|
| 237 |
+
"learning_rate": 9.817581034021272e-07,
|
| 238 |
+
"logits/chosen": 0.022985249757766724,
|
| 239 |
+
"logits/rejected": 0.06110963970422745,
|
| 240 |
+
"logps/chosen": -431.2767639160156,
|
| 241 |
+
"logps/rejected": -524.8790893554688,
|
| 242 |
+
"loss": 0.4341,
|
| 243 |
+
"rewards/accuracies": 0.699999988079071,
|
| 244 |
+
"rewards/chosen": -0.9490836262702942,
|
| 245 |
+
"rewards/margins": 1.221215009689331,
|
| 246 |
+
"rewards/rejected": -2.1702985763549805,
|
| 247 |
"step": 170
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.38,
|
| 251 |
+
"learning_rate": 9.765362502737097e-07,
|
| 252 |
+
"logits/chosen": -0.04698944836854935,
|
| 253 |
+
"logits/rejected": -0.01610407792031765,
|
| 254 |
+
"logps/chosen": -415.97393798828125,
|
| 255 |
+
"logps/rejected": -554.08837890625,
|
| 256 |
+
"loss": 0.4314,
|
| 257 |
+
"rewards/accuracies": 0.706250011920929,
|
| 258 |
+
"rewards/chosen": -0.8994191288948059,
|
| 259 |
+
"rewards/margins": 1.6427128314971924,
|
| 260 |
+
"rewards/rejected": -2.5421319007873535,
|
| 261 |
"step": 180
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.4,
|
| 265 |
+
"learning_rate": 9.706755858428485e-07,
|
| 266 |
+
"logits/chosen": 0.12246842682361603,
|
| 267 |
+
"logits/rejected": 0.13095493614673615,
|
| 268 |
+
"logps/chosen": -418.01751708984375,
|
| 269 |
+
"logps/rejected": -506.07421875,
|
| 270 |
+
"loss": 0.4235,
|
| 271 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 272 |
+
"rewards/chosen": -0.8335806131362915,
|
| 273 |
+
"rewards/margins": 1.0999118089675903,
|
| 274 |
+
"rewards/rejected": -1.9334923028945923,
|
| 275 |
"step": 190
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.42,
|
| 279 |
+
"learning_rate": 9.641839665080363e-07,
|
| 280 |
+
"logits/chosen": 0.11251389980316162,
|
| 281 |
+
"logits/rejected": 0.08771563321352005,
|
| 282 |
+
"logps/chosen": -444.196533203125,
|
| 283 |
+
"logps/rejected": -665.639892578125,
|
| 284 |
+
"loss": 0.419,
|
| 285 |
+
"rewards/accuracies": 0.75,
|
| 286 |
+
"rewards/chosen": -1.0370674133300781,
|
| 287 |
+
"rewards/margins": 2.6780648231506348,
|
| 288 |
+
"rewards/rejected": -3.7151317596435547,
|
| 289 |
"step": 200
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.44,
|
| 293 |
+
"learning_rate": 9.570700944819582e-07,
|
| 294 |
+
"logits/chosen": 0.16030333936214447,
|
| 295 |
+
"logits/rejected": 0.11628633737564087,
|
| 296 |
+
"logps/chosen": -400.63427734375,
|
| 297 |
+
"logps/rejected": -635.213134765625,
|
| 298 |
+
"loss": 0.4265,
|
| 299 |
+
"rewards/accuracies": 0.731249988079071,
|
| 300 |
+
"rewards/chosen": -0.8647702932357788,
|
| 301 |
+
"rewards/margins": 2.788623332977295,
|
| 302 |
+
"rewards/rejected": -3.6533939838409424,
|
| 303 |
"step": 210
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.46,
|
| 307 |
+
"learning_rate": 9.493435061259129e-07,
|
| 308 |
+
"logits/chosen": -0.053914181888103485,
|
| 309 |
+
"logits/rejected": -0.0958121120929718,
|
| 310 |
+
"logps/chosen": -501.2230529785156,
|
| 311 |
+
"logps/rejected": -713.2017211914062,
|
| 312 |
+
"loss": 0.4207,
|
| 313 |
+
"rewards/accuracies": 0.65625,
|
| 314 |
+
"rewards/chosen": -1.8454395532608032,
|
| 315 |
+
"rewards/margins": 2.649871826171875,
|
| 316 |
+
"rewards/rejected": -4.495311260223389,
|
| 317 |
"step": 220
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.48,
|
| 321 |
+
"learning_rate": 9.4101455916603e-07,
|
| 322 |
+
"logits/chosen": 0.024957913905382156,
|
| 323 |
+
"logits/rejected": -0.08018078655004501,
|
| 324 |
+
"logps/chosen": -582.2233276367188,
|
| 325 |
+
"logps/rejected": -891.3052978515625,
|
| 326 |
+
"loss": 0.4177,
|
| 327 |
+
"rewards/accuracies": 0.675000011920929,
|
| 328 |
+
"rewards/chosen": -2.561619758605957,
|
| 329 |
+
"rewards/margins": 3.5202267169952393,
|
| 330 |
+
"rewards/rejected": -6.081845760345459,
|
| 331 |
"step": 230
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 0.5,
|
| 335 |
+
"learning_rate": 9.320944188084241e-07,
|
| 336 |
+
"logits/chosen": 0.06781353056430817,
|
| 337 |
+
"logits/rejected": -0.011409667320549488,
|
| 338 |
+
"logps/chosen": -590.6744384765625,
|
| 339 |
+
"logps/rejected": -886.1702880859375,
|
| 340 |
+
"loss": 0.4203,
|
| 341 |
"rewards/accuracies": 0.6812499761581421,
|
| 342 |
+
"rewards/chosen": -2.787226915359497,
|
| 343 |
+
"rewards/margins": 3.05540132522583,
|
| 344 |
+
"rewards/rejected": -5.842628479003906,
|
| 345 |
"step": 240
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.52,
|
| 349 |
+
"learning_rate": 9.225950427718974e-07,
|
| 350 |
+
"logits/chosen": 0.17237909138202667,
|
| 351 |
+
"logits/rejected": 0.08455310761928558,
|
| 352 |
+
"logps/chosen": -487.7764587402344,
|
| 353 |
+
"logps/rejected": -983.6060791015625,
|
| 354 |
+
"loss": 0.4162,
|
| 355 |
+
"rewards/accuracies": 0.699999988079071,
|
| 356 |
+
"rewards/chosen": -1.7572829723358154,
|
| 357 |
+
"rewards/margins": 5.258338451385498,
|
| 358 |
+
"rewards/rejected": -7.015621185302734,
|
| 359 |
"step": 250
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.54,
|
| 363 |
+
"learning_rate": 9.125291652582547e-07,
|
| 364 |
+
"logits/chosen": 0.04219328239560127,
|
| 365 |
+
"logits/rejected": 0.006076293531805277,
|
| 366 |
+
"logps/chosen": -587.68212890625,
|
| 367 |
+
"logps/rejected": -869.4441528320312,
|
| 368 |
+
"loss": 0.4232,
|
| 369 |
+
"rewards/accuracies": 0.643750011920929,
|
| 370 |
+
"rewards/chosen": -2.330383777618408,
|
| 371 |
+
"rewards/margins": 3.363435745239258,
|
| 372 |
+
"rewards/rejected": -5.693819522857666,
|
| 373 |
"step": 260
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 0.57,
|
| 377 |
+
"learning_rate": 9.019102798817195e-07,
|
| 378 |
+
"logits/chosen": 0.1859409064054489,
|
| 379 |
+
"logits/rejected": -0.06210414692759514,
|
| 380 |
+
"logps/chosen": -533.7982177734375,
|
| 381 |
+
"logps/rejected": -985.5645751953125,
|
| 382 |
+
"loss": 0.4114,
|
| 383 |
"rewards/accuracies": 0.706250011920929,
|
| 384 |
+
"rewards/chosen": -2.0466434955596924,
|
| 385 |
+
"rewards/margins": 4.704514503479004,
|
| 386 |
+
"rewards/rejected": -6.751158237457275,
|
| 387 |
"step": 270
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 0.59,
|
| 391 |
+
"learning_rate": 8.90752621580335e-07,
|
| 392 |
+
"logits/chosen": 0.19852975010871887,
|
| 393 |
+
"logits/rejected": 0.030256235972046852,
|
| 394 |
+
"logps/chosen": -513.59033203125,
|
| 395 |
+
"logps/rejected": -937.9052734375,
|
| 396 |
+
"loss": 0.4156,
|
| 397 |
+
"rewards/accuracies": 0.6875,
|
| 398 |
+
"rewards/chosen": -2.1710736751556396,
|
| 399 |
+
"rewards/margins": 4.297308444976807,
|
| 400 |
+
"rewards/rejected": -6.468382358551025,
|
| 401 |
"step": 280
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.61,
|
| 405 |
+
"learning_rate": 8.79071147533597e-07,
|
| 406 |
+
"logits/chosen": 0.14935357868671417,
|
| 407 |
+
"logits/rejected": -0.028311902657151222,
|
| 408 |
+
"logps/chosen": -705.8804321289062,
|
| 409 |
+
"logps/rejected": -1405.2919921875,
|
| 410 |
+
"loss": 0.4052,
|
| 411 |
+
"rewards/accuracies": 0.706250011920929,
|
| 412 |
+
"rewards/chosen": -3.5944924354553223,
|
| 413 |
+
"rewards/margins": 7.230493068695068,
|
| 414 |
+
"rewards/rejected": -10.824986457824707,
|
| 415 |
"step": 290
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 0.63,
|
| 419 |
+
"learning_rate": 8.668815171119019e-07,
|
| 420 |
+
"logits/chosen": 0.2377752959728241,
|
| 421 |
+
"logits/rejected": 0.054682862013578415,
|
| 422 |
+
"logps/chosen": -492.9376525878906,
|
| 423 |
+
"logps/rejected": -1074.2073974609375,
|
| 424 |
+
"loss": 0.4129,
|
| 425 |
+
"rewards/accuracies": 0.762499988079071,
|
| 426 |
+
"rewards/chosen": -2.011826515197754,
|
| 427 |
+
"rewards/margins": 5.496823310852051,
|
| 428 |
+
"rewards/rejected": -7.5086493492126465,
|
| 429 |
"step": 300
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.65,
|
| 433 |
+
"learning_rate": 8.54200070884685e-07,
|
| 434 |
+
"logits/chosen": 0.2463679015636444,
|
| 435 |
+
"logits/rejected": -0.016352087259292603,
|
| 436 |
+
"logps/chosen": -525.9217529296875,
|
| 437 |
+
"logps/rejected": -1137.378662109375,
|
| 438 |
+
"loss": 0.4032,
|
| 439 |
+
"rewards/accuracies": 0.699999988079071,
|
| 440 |
+
"rewards/chosen": -2.3019001483917236,
|
| 441 |
+
"rewards/margins": 5.954804420471191,
|
| 442 |
+
"rewards/rejected": -8.25670337677002,
|
| 443 |
"step": 310
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.67,
|
| 447 |
+
"learning_rate": 8.410438087153911e-07,
|
| 448 |
+
"logits/chosen": 0.22406017780303955,
|
| 449 |
+
"logits/rejected": 0.10108964145183563,
|
| 450 |
+
"logps/chosen": -472.781005859375,
|
| 451 |
+
"logps/rejected": -1149.229248046875,
|
| 452 |
+
"loss": 0.4105,
|
| 453 |
+
"rewards/accuracies": 0.737500011920929,
|
| 454 |
+
"rewards/chosen": -1.571500301361084,
|
| 455 |
+
"rewards/margins": 7.018779754638672,
|
| 456 |
+
"rewards/rejected": -8.590279579162598,
|
| 457 |
"step": 320
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 0.69,
|
| 461 |
+
"learning_rate": 8.274303669726426e-07,
|
| 462 |
+
"logits/chosen": 0.29168057441711426,
|
| 463 |
+
"logits/rejected": 0.15196271240711212,
|
| 464 |
+
"logps/chosen": -600.2703857421875,
|
| 465 |
+
"logps/rejected": -1573.563232421875,
|
| 466 |
+
"loss": 0.402,
|
| 467 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 468 |
+
"rewards/chosen": -3.015018939971924,
|
| 469 |
+
"rewards/margins": 9.613102912902832,
|
| 470 |
+
"rewards/rejected": -12.628121376037598,
|
| 471 |
"step": 330
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 0.71,
|
| 475 |
+
"learning_rate": 8.133779948881513e-07,
|
| 476 |
+
"logits/chosen": 0.33964803814888,
|
| 477 |
+
"logits/rejected": 0.3003019690513611,
|
| 478 |
+
"logps/chosen": -494.3397521972656,
|
| 479 |
+
"logps/rejected": -1070.0369873046875,
|
| 480 |
+
"loss": 0.4047,
|
| 481 |
+
"rewards/accuracies": 0.800000011920929,
|
| 482 |
+
"rewards/chosen": -2.076432943344116,
|
| 483 |
+
"rewards/margins": 6.0156073570251465,
|
| 484 |
+
"rewards/rejected": -8.092041015625,
|
| 485 |
"step": 340
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.73,
|
| 489 |
+
"learning_rate": 7.989055300930704e-07,
|
| 490 |
+
"logits/chosen": 0.2937834858894348,
|
| 491 |
+
"logits/rejected": 0.22495004534721375,
|
| 492 |
+
"logps/chosen": -525.5234985351562,
|
| 493 |
+
"logps/rejected": -1205.3031005859375,
|
| 494 |
+
"loss": 0.404,
|
| 495 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 496 |
+
"rewards/chosen": -2.2402138710021973,
|
| 497 |
+
"rewards/margins": 6.707070350646973,
|
| 498 |
+
"rewards/rejected": -8.947283744812012,
|
| 499 |
"step": 350
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.75,
|
| 503 |
+
"learning_rate": 7.840323733655778e-07,
|
| 504 |
+
"logits/chosen": 0.20088157057762146,
|
| 505 |
+
"logits/rejected": 0.07263296842575073,
|
| 506 |
+
"logps/chosen": -569.1207275390625,
|
| 507 |
+
"logps/rejected": -1455.37548828125,
|
| 508 |
+
"loss": 0.3978,
|
| 509 |
+
"rewards/accuracies": 0.75,
|
| 510 |
+
"rewards/chosen": -2.4096343517303467,
|
| 511 |
+
"rewards/margins": 9.322293281555176,
|
| 512 |
+
"rewards/rejected": -11.731927871704102,
|
| 513 |
"step": 360
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 0.77,
|
| 517 |
+
"learning_rate": 7.687784626235447e-07,
|
| 518 |
+
"logits/chosen": 0.04894110560417175,
|
| 519 |
+
"logits/rejected": -0.0861678496003151,
|
| 520 |
+
"logps/chosen": -700.1285400390625,
|
| 521 |
+
"logps/rejected": -1508.282958984375,
|
| 522 |
+
"loss": 0.4002,
|
| 523 |
+
"rewards/accuracies": 0.699999988079071,
|
| 524 |
+
"rewards/chosen": -3.51697039604187,
|
| 525 |
+
"rewards/margins": 8.459676742553711,
|
| 526 |
+
"rewards/rejected": -11.97664737701416,
|
| 527 |
"step": 370
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 0.8,
|
| 531 |
+
"learning_rate": 7.531642461971514e-07,
|
| 532 |
+
"logits/chosen": 0.13316456973552704,
|
| 533 |
+
"logits/rejected": -0.1390618234872818,
|
| 534 |
+
"logps/chosen": -834.8723754882812,
|
| 535 |
+
"logps/rejected": -1582.9775390625,
|
| 536 |
+
"loss": 0.3995,
|
| 537 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 538 |
+
"rewards/chosen": -5.417990684509277,
|
| 539 |
+
"rewards/margins": 7.689940452575684,
|
| 540 |
+
"rewards/rejected": -13.107931137084961,
|
| 541 |
"step": 380
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 0.82,
|
| 545 |
+
"learning_rate": 7.372106554172801e-07,
|
| 546 |
+
"logits/chosen": 0.05666132643818855,
|
| 547 |
+
"logits/rejected": -0.015092259272933006,
|
| 548 |
+
"logps/chosen": -540.2429809570312,
|
| 549 |
+
"logps/rejected": -1435.7958984375,
|
| 550 |
+
"loss": 0.3973,
|
| 551 |
+
"rewards/accuracies": 0.71875,
|
| 552 |
+
"rewards/chosen": -1.985515832901001,
|
| 553 |
+
"rewards/margins": 9.362586975097656,
|
| 554 |
+
"rewards/rejected": -11.348101615905762,
|
| 555 |
"step": 390
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 0.84,
|
| 559 |
+
"learning_rate": 7.209390765564318e-07,
|
| 560 |
+
"logits/chosen": 0.16617372632026672,
|
| 561 |
+
"logits/rejected": -0.0461905337870121,
|
| 562 |
+
"logps/chosen": -723.858642578125,
|
| 563 |
+
"logps/rejected": -1655.4898681640625,
|
| 564 |
+
"loss": 0.4013,
|
| 565 |
+
"rewards/accuracies": 0.737500011920929,
|
| 566 |
+
"rewards/chosen": -4.068535804748535,
|
| 567 |
+
"rewards/margins": 9.538484573364258,
|
| 568 |
+
"rewards/rejected": -13.607022285461426,
|
| 569 |
"step": 400
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.86,
|
| 573 |
+
"learning_rate": 7.043713221597773e-07,
|
| 574 |
+
"logits/chosen": 0.06812890619039536,
|
| 575 |
+
"logits/rejected": -0.07965459674596786,
|
| 576 |
+
"logps/chosen": -683.1234130859375,
|
| 577 |
+
"logps/rejected": -1550.495361328125,
|
| 578 |
+
"loss": 0.4009,
|
| 579 |
+
"rewards/accuracies": 0.6812499761581421,
|
| 580 |
+
"rewards/chosen": -3.939258575439453,
|
| 581 |
+
"rewards/margins": 8.659662246704102,
|
| 582 |
+
"rewards/rejected": -12.598922729492188,
|
| 583 |
"step": 410
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 0.88,
|
| 587 |
+
"learning_rate": 6.875296018047809e-07,
|
| 588 |
+
"logits/chosen": 0.20702295005321503,
|
| 589 |
+
"logits/rejected": 0.037614382803440094,
|
| 590 |
+
"logps/chosen": -662.0958251953125,
|
| 591 |
+
"logps/rejected": -1592.722900390625,
|
| 592 |
+
"loss": 0.3984,
|
| 593 |
+
"rewards/accuracies": 0.6875,
|
| 594 |
+
"rewards/chosen": -3.6876208782196045,
|
| 595 |
+
"rewards/margins": 9.383062362670898,
|
| 596 |
+
"rewards/rejected": -13.070683479309082,
|
| 597 |
"step": 420
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 0.9,
|
| 601 |
+
"learning_rate": 6.704364923285857e-07,
|
| 602 |
+
"logits/chosen": 0.0521254763007164,
|
| 603 |
+
"logits/rejected": -0.16673153638839722,
|
| 604 |
+
"logps/chosen": -708.7855224609375,
|
| 605 |
+
"logps/rejected": -1829.0189208984375,
|
| 606 |
+
"loss": 0.398,
|
| 607 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 608 |
+
"rewards/chosen": -3.7470862865448,
|
| 609 |
+
"rewards/margins": 11.305286407470703,
|
| 610 |
+
"rewards/rejected": -15.052372932434082,
|
| 611 |
"step": 430
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 0.92,
|
| 615 |
+
"learning_rate": 6.531149075630796e-07,
|
| 616 |
+
"logits/chosen": 0.03881196305155754,
|
| 617 |
+
"logits/rejected": -0.04577777534723282,
|
| 618 |
+
"logps/chosen": -575.1758422851562,
|
| 619 |
+
"logps/rejected": -1710.799072265625,
|
| 620 |
+
"loss": 0.399,
|
| 621 |
+
"rewards/accuracies": 0.706250011920929,
|
| 622 |
+
"rewards/chosen": -2.906187057495117,
|
| 623 |
+
"rewards/margins": 11.424031257629395,
|
| 624 |
+
"rewards/rejected": -14.330218315124512,
|
| 625 |
"step": 440
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 0.94,
|
| 629 |
+
"learning_rate": 6.355880676182085e-07,
|
| 630 |
+
"logits/chosen": 0.03127314895391464,
|
| 631 |
+
"logits/rejected": -0.32143911719322205,
|
| 632 |
+
"logps/chosen": -950.0020751953125,
|
| 633 |
+
"logps/rejected": -2152.05029296875,
|
| 634 |
+
"loss": 0.3881,
|
| 635 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 636 |
+
"rewards/chosen": -6.0005059242248535,
|
| 637 |
+
"rewards/margins": 12.68049144744873,
|
| 638 |
+
"rewards/rejected": -18.68099594116211,
|
| 639 |
"step": 450
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.96,
|
| 643 |
+
"learning_rate": 6.178794677547137e-07,
|
| 644 |
+
"logits/chosen": 0.10744090378284454,
|
| 645 |
+
"logits/rejected": -0.19111321866512299,
|
| 646 |
+
"logps/chosen": -911.0108642578125,
|
| 647 |
+
"logps/rejected": -2054.70556640625,
|
| 648 |
+
"loss": 0.3996,
|
| 649 |
+
"rewards/accuracies": 0.668749988079071,
|
| 650 |
+
"rewards/chosen": -6.130732536315918,
|
| 651 |
+
"rewards/margins": 11.758760452270508,
|
| 652 |
+
"rewards/rejected": -17.88949203491211,
|
| 653 |
"step": 460
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 0.98,
|
| 657 |
+
"learning_rate": 6.000128468880222e-07,
|
| 658 |
+
"logits/chosen": -0.006098261568695307,
|
| 659 |
+
"logits/rejected": -0.45566266775131226,
|
| 660 |
+
"logps/chosen": -1054.670166015625,
|
| 661 |
+
"logps/rejected": -2545.79296875,
|
| 662 |
+
"loss": 0.3969,
|
| 663 |
+
"rewards/accuracies": 0.78125,
|
| 664 |
+
"rewards/chosen": -7.155210018157959,
|
| 665 |
+
"rewards/margins": 15.32789134979248,
|
| 666 |
+
"rewards/rejected": -22.48310089111328,
|
| 667 |
"step": 470
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 1.0,
|
| 671 |
+
"learning_rate": 5.820121557655108e-07,
|
| 672 |
+
"logits/chosen": 0.051991622895002365,
|
| 673 |
+
"logits/rejected": -0.2006780356168747,
|
| 674 |
+
"logps/chosen": -758.8955078125,
|
| 675 |
+
"logps/rejected": -2237.24267578125,
|
| 676 |
+
"loss": 0.3949,
|
| 677 |
+
"rewards/accuracies": 0.78125,
|
| 678 |
+
"rewards/chosen": -4.22313928604126,
|
| 679 |
+
"rewards/margins": 14.878347396850586,
|
| 680 |
+
"rewards/rejected": -19.101486206054688,
|
| 681 |
"step": 480
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 1.03,
|
| 685 |
+
"learning_rate": 5.639015248598023e-07,
|
| 686 |
+
"logits/chosen": -0.026292938739061356,
|
| 687 |
+
"logits/rejected": -0.274114191532135,
|
| 688 |
+
"logps/chosen": -778.0355224609375,
|
| 689 |
+
"logps/rejected": -1928.9573974609375,
|
| 690 |
+
"loss": 0.3625,
|
| 691 |
+
"rewards/accuracies": 0.7875000238418579,
|
| 692 |
+
"rewards/chosen": -4.59261417388916,
|
| 693 |
+
"rewards/margins": 11.645755767822266,
|
| 694 |
+
"rewards/rejected": -16.23836898803711,
|
| 695 |
"step": 490
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 1.05,
|
| 699 |
+
"learning_rate": 5.457052320211339e-07,
|
| 700 |
+
"logits/chosen": -0.017996052280068398,
|
| 701 |
+
"logits/rejected": -0.45051470398902893,
|
| 702 |
+
"logps/chosen": -968.8033447265625,
|
| 703 |
+
"logps/rejected": -2387.48486328125,
|
| 704 |
+
"loss": 0.3659,
|
| 705 |
"rewards/accuracies": 0.762499988079071,
|
| 706 |
+
"rewards/chosen": -6.734767913818359,
|
| 707 |
+
"rewards/margins": 14.207613945007324,
|
| 708 |
+
"rewards/rejected": -20.942384719848633,
|
| 709 |
"step": 500
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 1.07,
|
| 713 |
+
"learning_rate": 5.274476699321637e-07,
|
| 714 |
+
"logits/chosen": 0.005720546934753656,
|
| 715 |
+
"logits/rejected": -0.5387086868286133,
|
| 716 |
+
"logps/chosen": -1205.0419921875,
|
| 717 |
+
"logps/rejected": -2745.4345703125,
|
| 718 |
+
"loss": 0.3687,
|
| 719 |
+
"rewards/accuracies": 0.8125,
|
| 720 |
+
"rewards/chosen": -8.98927116394043,
|
| 721 |
+
"rewards/margins": 15.691691398620605,
|
| 722 |
+
"rewards/rejected": -24.680959701538086,
|
| 723 |
"step": 510
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 1.09,
|
| 727 |
+
"learning_rate": 5.091533134088387e-07,
|
| 728 |
+
"logits/chosen": -0.005288724787533283,
|
| 729 |
+
"logits/rejected": -0.1951751857995987,
|
| 730 |
+
"logps/chosen": -1033.7554931640625,
|
| 731 |
+
"logps/rejected": -2083.949462890625,
|
| 732 |
+
"loss": 0.3716,
|
| 733 |
+
"rewards/accuracies": 0.737500011920929,
|
| 734 |
+
"rewards/chosen": -6.909409523010254,
|
| 735 |
+
"rewards/margins": 10.849687576293945,
|
| 736 |
+
"rewards/rejected": -17.75909423828125,
|
| 737 |
"step": 520
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 1.11,
|
| 741 |
+
"learning_rate": 4.908466865911614e-07,
|
| 742 |
+
"logits/chosen": 0.07674840837717056,
|
| 743 |
+
"logits/rejected": -0.22934529185295105,
|
| 744 |
+
"logps/chosen": -611.5888671875,
|
| 745 |
+
"logps/rejected": -1846.804443359375,
|
| 746 |
+
"loss": 0.3658,
|
| 747 |
+
"rewards/accuracies": 0.762499988079071,
|
| 748 |
+
"rewards/chosen": -2.9781932830810547,
|
| 749 |
+
"rewards/margins": 12.689587593078613,
|
| 750 |
+
"rewards/rejected": -15.667780876159668,
|
| 751 |
"step": 530
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 1.13,
|
| 755 |
+
"learning_rate": 4.7255233006783624e-07,
|
| 756 |
+
"logits/chosen": 0.021334605291485786,
|
| 757 |
+
"logits/rejected": -0.3709332048892975,
|
| 758 |
+
"logps/chosen": -970.0914306640625,
|
| 759 |
+
"logps/rejected": -2093.108642578125,
|
| 760 |
+
"loss": 0.3666,
|
| 761 |
+
"rewards/accuracies": 0.75,
|
| 762 |
+
"rewards/chosen": -6.791567802429199,
|
| 763 |
+
"rewards/margins": 11.35092544555664,
|
| 764 |
+
"rewards/rejected": -18.142492294311523,
|
| 765 |
"step": 540
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 1.15,
|
| 769 |
+
"learning_rate": 4.5429476797886617e-07,
|
| 770 |
+
"logits/chosen": 0.033512182533741,
|
| 771 |
+
"logits/rejected": -0.3936399519443512,
|
| 772 |
+
"logps/chosen": -856.0861206054688,
|
| 773 |
+
"logps/rejected": -2535.446044921875,
|
| 774 |
+
"loss": 0.3693,
|
| 775 |
+
"rewards/accuracies": 0.75,
|
| 776 |
+
"rewards/chosen": -5.2735795974731445,
|
| 777 |
+
"rewards/margins": 17.031352996826172,
|
| 778 |
+
"rewards/rejected": -22.304935455322266,
|
| 779 |
"step": 550
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 1.17,
|
| 783 |
+
"learning_rate": 4.3609847514019763e-07,
|
| 784 |
+
"logits/chosen": -0.05847325176000595,
|
| 785 |
+
"logits/rejected": -0.5101506114006042,
|
| 786 |
+
"logps/chosen": -715.2965087890625,
|
| 787 |
+
"logps/rejected": -1808.1195068359375,
|
| 788 |
+
"loss": 0.3829,
|
| 789 |
"rewards/accuracies": 0.793749988079071,
|
| 790 |
+
"rewards/chosen": -3.906663179397583,
|
| 791 |
+
"rewards/margins": 11.054914474487305,
|
| 792 |
+
"rewards/rejected": -14.961578369140625,
|
| 793 |
"step": 560
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"epoch": 1.19,
|
| 797 |
+
"learning_rate": 4.179878442344892e-07,
|
| 798 |
+
"logits/chosen": 0.03838271647691727,
|
| 799 |
+
"logits/rejected": -0.41066282987594604,
|
| 800 |
+
"logps/chosen": -600.6312866210938,
|
| 801 |
+
"logps/rejected": -1975.118408203125,
|
| 802 |
+
"loss": 0.3707,
|
| 803 |
+
"rewards/accuracies": 0.800000011920929,
|
| 804 |
+
"rewards/chosen": -3.174193859100342,
|
| 805 |
+
"rewards/margins": 13.602760314941406,
|
| 806 |
+
"rewards/rejected": -16.77695655822754,
|
| 807 |
"step": 570
|
| 808 |
},
|
| 809 |
{
|
| 810 |
+
"epoch": 1.21,
|
| 811 |
+
"learning_rate": 3.9998715311197783e-07,
|
| 812 |
+
"logits/chosen": 0.00818496011197567,
|
| 813 |
+
"logits/rejected": -0.46959954500198364,
|
| 814 |
+
"logps/chosen": -1030.3143310546875,
|
| 815 |
+
"logps/rejected": -2606.57666015625,
|
| 816 |
+
"loss": 0.3699,
|
| 817 |
+
"rewards/accuracies": 0.768750011920929,
|
| 818 |
+
"rewards/chosen": -7.15012264251709,
|
| 819 |
+
"rewards/margins": 15.829042434692383,
|
| 820 |
+
"rewards/rejected": -22.979164123535156,
|
| 821 |
"step": 580
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 1.24,
|
| 825 |
+
"learning_rate": 3.821205322452863e-07,
|
| 826 |
+
"logits/chosen": 0.0865226536989212,
|
| 827 |
+
"logits/rejected": -0.49088770151138306,
|
| 828 |
+
"logps/chosen": -979.1468505859375,
|
| 829 |
+
"logps/rejected": -2228.843017578125,
|
| 830 |
+
"loss": 0.3649,
|
| 831 |
+
"rewards/accuracies": 0.737500011920929,
|
| 832 |
+
"rewards/chosen": -6.757370948791504,
|
| 833 |
+
"rewards/margins": 12.640668869018555,
|
| 834 |
+
"rewards/rejected": -19.398040771484375,
|
| 835 |
"step": 590
|
| 836 |
},
|
| 837 |
{
|
| 838 |
+
"epoch": 1.26,
|
| 839 |
+
"learning_rate": 3.6441193238179146e-07,
|
| 840 |
+
"logits/chosen": 0.06730663031339645,
|
| 841 |
+
"logits/rejected": -0.47154170274734497,
|
| 842 |
+
"logps/chosen": -628.3743896484375,
|
| 843 |
+
"logps/rejected": -2107.99169921875,
|
| 844 |
+
"loss": 0.359,
|
| 845 |
+
"rewards/accuracies": 0.824999988079071,
|
| 846 |
+
"rewards/chosen": -3.383096218109131,
|
| 847 |
+
"rewards/margins": 14.71300983428955,
|
| 848 |
+
"rewards/rejected": -18.096105575561523,
|
| 849 |
"step": 600
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 1.28,
|
| 853 |
+
"learning_rate": 3.4688509243692034e-07,
|
| 854 |
+
"logits/chosen": -0.07551614940166473,
|
| 855 |
+
"logits/rejected": -0.4228256344795227,
|
| 856 |
+
"logps/chosen": -625.5606689453125,
|
| 857 |
+
"logps/rejected": -2476.503173828125,
|
| 858 |
+
"loss": 0.3736,
|
| 859 |
+
"rewards/accuracies": 0.768750011920929,
|
| 860 |
+
"rewards/chosen": -3.288256883621216,
|
| 861 |
+
"rewards/margins": 17.969881057739258,
|
| 862 |
+
"rewards/rejected": -21.258136749267578,
|
| 863 |
"step": 610
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 1.3,
|
| 867 |
+
"learning_rate": 3.295635076714144e-07,
|
| 868 |
+
"logits/chosen": -0.012427730485796928,
|
| 869 |
+
"logits/rejected": -0.44272977113723755,
|
| 870 |
+
"logps/chosen": -558.3466796875,
|
| 871 |
+
"logps/rejected": -1993.5582275390625,
|
| 872 |
+
"loss": 0.3655,
|
| 873 |
+
"rewards/accuracies": 0.8062499761581421,
|
| 874 |
+
"rewards/chosen": -3.1298646926879883,
|
| 875 |
+
"rewards/margins": 14.332046508789062,
|
| 876 |
+
"rewards/rejected": -17.461912155151367,
|
| 877 |
"step": 620
|
| 878 |
},
|
| 879 |
{
|
| 880 |
+
"epoch": 1.32,
|
| 881 |
+
"learning_rate": 3.12470398195219e-07,
|
| 882 |
+
"logits/chosen": -0.04589563235640526,
|
| 883 |
+
"logits/rejected": -0.566718339920044,
|
| 884 |
+
"logps/chosen": -788.1644287109375,
|
| 885 |
+
"logps/rejected": -2312.9814453125,
|
| 886 |
+
"loss": 0.3652,
|
| 887 |
+
"rewards/accuracies": 0.75,
|
| 888 |
+
"rewards/chosen": -4.818960189819336,
|
| 889 |
+
"rewards/margins": 14.982978820800781,
|
| 890 |
+
"rewards/rejected": -19.80194091796875,
|
| 891 |
"step": 630
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 1.34,
|
| 895 |
+
"learning_rate": 2.956286778402226e-07,
|
| 896 |
+
"logits/chosen": -0.10815076529979706,
|
| 897 |
+
"logits/rejected": -0.40516456961631775,
|
| 898 |
+
"logps/chosen": -773.5537719726562,
|
| 899 |
+
"logps/rejected": -2297.0078125,
|
| 900 |
+
"loss": 0.3614,
|
| 901 |
+
"rewards/accuracies": 0.800000011920929,
|
| 902 |
+
"rewards/chosen": -3.9812560081481934,
|
| 903 |
+
"rewards/margins": 15.931207656860352,
|
| 904 |
+
"rewards/rejected": -19.912464141845703,
|
| 905 |
"step": 640
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 1.36,
|
| 909 |
+
"learning_rate": 2.7906092344356826e-07,
|
| 910 |
+
"logits/chosen": 0.0947767049074173,
|
| 911 |
+
"logits/rejected": -0.34178268909454346,
|
| 912 |
+
"logps/chosen": -563.9888916015625,
|
| 913 |
+
"logps/rejected": -1955.478759765625,
|
| 914 |
+
"loss": 0.3597,
|
| 915 |
+
"rewards/accuracies": 0.8062499761581421,
|
| 916 |
+
"rewards/chosen": -2.7671079635620117,
|
| 917 |
+
"rewards/margins": 14.173640251159668,
|
| 918 |
+
"rewards/rejected": -16.940750122070312,
|
| 919 |
"step": 650
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 1.38,
|
| 923 |
+
"learning_rate": 2.6278934458271996e-07,
|
| 924 |
+
"logits/chosen": -0.14527355134487152,
|
| 925 |
+
"logits/rejected": -0.5815786123275757,
|
| 926 |
+
"logps/chosen": -862.9083862304688,
|
| 927 |
+
"logps/rejected": -2731.31298828125,
|
| 928 |
+
"loss": 0.3611,
|
| 929 |
+
"rewards/accuracies": 0.762499988079071,
|
| 930 |
+
"rewards/chosen": -5.637027740478516,
|
| 931 |
+
"rewards/margins": 18.794246673583984,
|
| 932 |
+
"rewards/rejected": -24.4312744140625,
|
| 933 |
"step": 660
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 1.4,
|
| 937 |
+
"learning_rate": 2.468357538028487e-07,
|
| 938 |
+
"logits/chosen": -0.026713650673627853,
|
| 939 |
+
"logits/rejected": -0.5581346750259399,
|
| 940 |
+
"logps/chosen": -826.4105224609375,
|
| 941 |
+
"logps/rejected": -2481.40087890625,
|
| 942 |
+
"loss": 0.3602,
|
| 943 |
"rewards/accuracies": 0.768750011920929,
|
| 944 |
+
"rewards/chosen": -5.318304061889648,
|
| 945 |
+
"rewards/margins": 16.639951705932617,
|
| 946 |
+
"rewards/rejected": -21.958255767822266,
|
| 947 |
"step": 670
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 1.42,
|
| 951 |
+
"learning_rate": 2.312215373764551e-07,
|
| 952 |
+
"logits/chosen": -0.17602002620697021,
|
| 953 |
+
"logits/rejected": -0.7885143756866455,
|
| 954 |
+
"logps/chosen": -1041.0810546875,
|
| 955 |
+
"logps/rejected": -2838.9697265625,
|
| 956 |
+
"loss": 0.3635,
|
| 957 |
+
"rewards/accuracies": 0.78125,
|
| 958 |
+
"rewards/chosen": -6.526537895202637,
|
| 959 |
+
"rewards/margins": 18.3980655670166,
|
| 960 |
+
"rewards/rejected": -24.924602508544922,
|
| 961 |
"step": 680
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"epoch": 1.44,
|
| 965 |
+
"learning_rate": 2.1596762663442213e-07,
|
| 966 |
+
"logits/chosen": -0.11598227918148041,
|
| 967 |
+
"logits/rejected": -0.8071243166923523,
|
| 968 |
+
"logps/chosen": -812.7769165039062,
|
| 969 |
+
"logps/rejected": -2409.599365234375,
|
| 970 |
+
"loss": 0.3651,
|
| 971 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 972 |
+
"rewards/chosen": -5.32067346572876,
|
| 973 |
+
"rewards/margins": 16.22947120666504,
|
| 974 |
+
"rewards/rejected": -21.55014419555664,
|
| 975 |
"step": 690
|
| 976 |
},
|
| 977 |
{
|
| 978 |
+
"epoch": 1.47,
|
| 979 |
+
"learning_rate": 2.0109446990692963e-07,
|
| 980 |
+
"logits/chosen": -0.1322178840637207,
|
| 981 |
+
"logits/rejected": -0.6750038862228394,
|
| 982 |
+
"logps/chosen": -735.38623046875,
|
| 983 |
+
"logps/rejected": -2221.10107421875,
|
| 984 |
+
"loss": 0.3629,
|
| 985 |
+
"rewards/accuracies": 0.793749988079071,
|
| 986 |
+
"rewards/chosen": -4.028614044189453,
|
| 987 |
+
"rewards/margins": 14.953264236450195,
|
| 988 |
+
"rewards/rejected": -18.98187828063965,
|
| 989 |
"step": 700
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 1.49,
|
| 993 |
+
"learning_rate": 1.8662200511184872e-07,
|
| 994 |
+
"logits/chosen": -0.07118358463048935,
|
| 995 |
+
"logits/rejected": -0.6492079496383667,
|
| 996 |
+
"logps/chosen": -733.4075317382812,
|
| 997 |
+
"logps/rejected": -1978.872314453125,
|
| 998 |
+
"loss": 0.3605,
|
| 999 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 1000 |
+
"rewards/chosen": -4.418785572052002,
|
| 1001 |
+
"rewards/margins": 12.647371292114258,
|
| 1002 |
+
"rewards/rejected": -17.0661563873291,
|
| 1003 |
"step": 710
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
+
"epoch": 1.51,
|
| 1007 |
+
"learning_rate": 1.725696330273575e-07,
|
| 1008 |
+
"logits/chosen": -0.013000762090086937,
|
| 1009 |
+
"logits/rejected": -0.4818207621574402,
|
| 1010 |
+
"logps/chosen": -668.6626586914062,
|
| 1011 |
+
"logps/rejected": -2029.7943115234375,
|
| 1012 |
+
"loss": 0.3526,
|
| 1013 |
+
"rewards/accuracies": 0.8125,
|
| 1014 |
+
"rewards/chosen": -3.4144835472106934,
|
| 1015 |
+
"rewards/margins": 13.840230941772461,
|
| 1016 |
+
"rewards/rejected": -17.25471305847168,
|
| 1017 |
"step": 720
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
+
"epoch": 1.53,
|
| 1021 |
+
"learning_rate": 1.589561912846089e-07,
|
| 1022 |
+
"logits/chosen": -0.08404045552015305,
|
| 1023 |
+
"logits/rejected": -0.6345758438110352,
|
| 1024 |
+
"logps/chosen": -747.2855224609375,
|
| 1025 |
+
"logps/rejected": -2472.15234375,
|
| 1026 |
+
"loss": 0.3523,
|
| 1027 |
+
"rewards/accuracies": 0.8125,
|
| 1028 |
+
"rewards/chosen": -4.540907859802246,
|
| 1029 |
+
"rewards/margins": 17.551443099975586,
|
| 1030 |
+
"rewards/rejected": -22.092350006103516,
|
| 1031 |
"step": 730
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
+
"epoch": 1.55,
|
| 1035 |
+
"learning_rate": 1.4579992911531496e-07,
|
| 1036 |
+
"logits/chosen": -0.21502713859081268,
|
| 1037 |
+
"logits/rejected": -0.8310446739196777,
|
| 1038 |
+
"logps/chosen": -1271.535888671875,
|
| 1039 |
+
"logps/rejected": -2498.06884765625,
|
| 1040 |
+
"loss": 0.3595,
|
| 1041 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 1042 |
+
"rewards/chosen": -9.246122360229492,
|
| 1043 |
+
"rewards/margins": 12.74262809753418,
|
| 1044 |
+
"rewards/rejected": -21.988750457763672,
|
| 1045 |
"step": 740
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
+
"epoch": 1.57,
|
| 1049 |
+
"learning_rate": 1.3311848288809813e-07,
|
| 1050 |
+
"logits/chosen": -0.06439349055290222,
|
| 1051 |
+
"logits/rejected": -0.5617343187332153,
|
| 1052 |
+
"logps/chosen": -864.2130126953125,
|
| 1053 |
+
"logps/rejected": -2242.72119140625,
|
| 1054 |
+
"loss": 0.3647,
|
| 1055 |
+
"rewards/accuracies": 0.768750011920929,
|
| 1056 |
+
"rewards/chosen": -5.325442790985107,
|
| 1057 |
+
"rewards/margins": 14.041879653930664,
|
| 1058 |
+
"rewards/rejected": -19.367321014404297,
|
| 1059 |
"step": 750
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
+
"epoch": 1.59,
|
| 1063 |
+
"learning_rate": 1.209288524664029e-07,
|
| 1064 |
+
"logits/chosen": -0.007544988300651312,
|
| 1065 |
+
"logits/rejected": -0.6543707251548767,
|
| 1066 |
+
"logps/chosen": -929.5089721679688,
|
| 1067 |
+
"logps/rejected": -2724.5986328125,
|
| 1068 |
+
"loss": 0.3608,
|
| 1069 |
+
"rewards/accuracies": 0.768750011920929,
|
| 1070 |
+
"rewards/chosen": -5.293430805206299,
|
| 1071 |
+
"rewards/margins": 18.260942459106445,
|
| 1072 |
+
"rewards/rejected": -23.554372787475586,
|
| 1073 |
"step": 760
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"epoch": 1.61,
|
| 1077 |
+
"learning_rate": 1.0924737841966497e-07,
|
| 1078 |
+
"logits/chosen": -0.09723073244094849,
|
| 1079 |
+
"logits/rejected": -0.6276915669441223,
|
| 1080 |
+
"logps/chosen": -1004.7063598632812,
|
| 1081 |
+
"logps/rejected": -2738.22119140625,
|
| 1082 |
+
"loss": 0.3611,
|
| 1083 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 1084 |
+
"rewards/chosen": -6.353689193725586,
|
| 1085 |
+
"rewards/margins": 17.72109031677246,
|
| 1086 |
+
"rewards/rejected": -24.074779510498047,
|
| 1087 |
"step": 770
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
+
"epoch": 1.63,
|
| 1091 |
+
"learning_rate": 9.808972011828054e-08,
|
| 1092 |
+
"logits/chosen": -0.10224993526935577,
|
| 1093 |
+
"logits/rejected": -0.7376490831375122,
|
| 1094 |
+
"logps/chosen": -815.9857788085938,
|
| 1095 |
+
"logps/rejected": -2558.365478515625,
|
| 1096 |
+
"loss": 0.3546,
|
| 1097 |
+
"rewards/accuracies": 0.78125,
|
| 1098 |
+
"rewards/chosen": -5.3413801193237305,
|
| 1099 |
+
"rewards/margins": 17.04741859436035,
|
| 1100 |
+
"rewards/rejected": -22.388797760009766,
|
| 1101 |
"step": 780
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
+
"epoch": 1.65,
|
| 1105 |
+
"learning_rate": 8.747083474174527e-08,
|
| 1106 |
+
"logits/chosen": -0.06878294795751572,
|
| 1107 |
+
"logits/rejected": -0.6381550431251526,
|
| 1108 |
+
"logps/chosen": -684.91455078125,
|
| 1109 |
+
"logps/rejected": -2131.619140625,
|
| 1110 |
+
"loss": 0.3659,
|
| 1111 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1112 |
+
"rewards/chosen": -3.895407199859619,
|
| 1113 |
+
"rewards/margins": 14.6422700881958,
|
| 1114 |
+
"rewards/rejected": -18.537677764892578,
|
| 1115 |
"step": 790
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
+
"epoch": 1.67,
|
| 1119 |
+
"learning_rate": 7.740495722810269e-08,
|
| 1120 |
+
"logits/chosen": -0.13722161948680878,
|
| 1121 |
+
"logits/rejected": -0.8848565220832825,
|
| 1122 |
+
"logps/chosen": -734.0003662109375,
|
| 1123 |
+
"logps/rejected": -2482.77490234375,
|
| 1124 |
+
"loss": 0.3646,
|
| 1125 |
+
"rewards/accuracies": 0.78125,
|
| 1126 |
+
"rewards/chosen": -4.181973934173584,
|
| 1127 |
+
"rewards/margins": 17.644306182861328,
|
| 1128 |
+
"rewards/rejected": -21.826278686523438,
|
| 1129 |
"step": 800
|
| 1130 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1131 |
{
|
| 1132 |
"epoch": 1.7,
|
| 1133 |
+
"learning_rate": 6.790558119157597e-08,
|
| 1134 |
+
"logits/chosen": -0.0752606987953186,
|
| 1135 |
+
"logits/rejected": -0.6523066163063049,
|
| 1136 |
+
"logps/chosen": -913.1563720703125,
|
| 1137 |
+
"logps/rejected": -2458.097412109375,
|
| 1138 |
+
"loss": 0.3615,
|
| 1139 |
+
"rewards/accuracies": 0.78125,
|
| 1140 |
+
"rewards/chosen": -5.732936859130859,
|
| 1141 |
+
"rewards/margins": 15.887182235717773,
|
| 1142 |
+
"rewards/rejected": -21.620121002197266,
|
| 1143 |
+
"step": 810
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 1.72,
|
| 1147 |
+
"learning_rate": 5.898544083397e-08,
|
| 1148 |
+
"logits/chosen": 0.020809601992368698,
|
| 1149 |
+
"logits/rejected": -0.5849136114120483,
|
| 1150 |
+
"logps/chosen": -540.7110595703125,
|
| 1151 |
+
"logps/rejected": -2217.325927734375,
|
| 1152 |
+
"loss": 0.3579,
|
| 1153 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 1154 |
+
"rewards/chosen": -2.4081716537475586,
|
| 1155 |
+
"rewards/margins": 16.857872009277344,
|
| 1156 |
+
"rewards/rejected": -19.266042709350586,
|
| 1157 |
+
"step": 820
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 1.74,
|
| 1161 |
+
"learning_rate": 5.065649387408705e-08,
|
| 1162 |
+
"logits/chosen": -0.09206173568964005,
|
| 1163 |
+
"logits/rejected": -0.7031819224357605,
|
| 1164 |
+
"logps/chosen": -1010.9982299804688,
|
| 1165 |
+
"logps/rejected": -2271.75537109375,
|
| 1166 |
+
"loss": 0.3578,
|
| 1167 |
+
"rewards/accuracies": 0.7562500238418579,
|
| 1168 |
+
"rewards/chosen": -6.834753513336182,
|
| 1169 |
+
"rewards/margins": 12.86634349822998,
|
| 1170 |
+
"rewards/rejected": -19.701095581054688,
|
| 1171 |
+
"step": 830
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
"epoch": 1.76,
|
| 1175 |
+
"learning_rate": 4.292990551804171e-08,
|
| 1176 |
+
"logits/chosen": -0.19090762734413147,
|
| 1177 |
+
"logits/rejected": -0.7296367883682251,
|
| 1178 |
+
"logps/chosen": -932.6066284179688,
|
| 1179 |
+
"logps/rejected": -2595.78759765625,
|
| 1180 |
+
"loss": 0.3589,
|
| 1181 |
+
"rewards/accuracies": 0.793749988079071,
|
| 1182 |
+
"rewards/chosen": -6.08009672164917,
|
| 1183 |
+
"rewards/margins": 16.833927154541016,
|
| 1184 |
+
"rewards/rejected": -22.914024353027344,
|
| 1185 |
+
"step": 840
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
"epoch": 1.78,
|
| 1189 |
+
"learning_rate": 3.581603349196371e-08,
|
| 1190 |
+
"logits/chosen": -0.07513806223869324,
|
| 1191 |
+
"logits/rejected": -0.844355583190918,
|
| 1192 |
+
"logps/chosen": -655.0567626953125,
|
| 1193 |
+
"logps/rejected": -2801.64501953125,
|
| 1194 |
+
"loss": 0.3561,
|
| 1195 |
+
"rewards/accuracies": 0.8125,
|
| 1196 |
+
"rewards/chosen": -3.282649517059326,
|
| 1197 |
+
"rewards/margins": 21.719741821289062,
|
| 1198 |
+
"rewards/rejected": -25.002391815185547,
|
| 1199 |
+
"step": 850
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
"epoch": 1.8,
|
| 1203 |
+
"learning_rate": 2.9324414157151367e-08,
|
| 1204 |
+
"logits/chosen": -0.12824216485023499,
|
| 1205 |
+
"logits/rejected": -0.7398759126663208,
|
| 1206 |
+
"logps/chosen": -818.0006103515625,
|
| 1207 |
+
"logps/rejected": -2734.663818359375,
|
| 1208 |
+
"loss": 0.3529,
|
| 1209 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 1210 |
+
"rewards/chosen": -4.799277305603027,
|
| 1211 |
+
"rewards/margins": 19.707080841064453,
|
| 1212 |
+
"rewards/rejected": -24.506359100341797,
|
| 1213 |
+
"step": 860
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
"epoch": 1.82,
|
| 1217 |
+
"learning_rate": 2.3463749726290284e-08,
|
| 1218 |
+
"logits/chosen": -0.22140809893608093,
|
| 1219 |
+
"logits/rejected": -0.7506135702133179,
|
| 1220 |
+
"logps/chosen": -939.72900390625,
|
| 1221 |
+
"logps/rejected": -2747.75537109375,
|
| 1222 |
+
"loss": 0.3642,
|
| 1223 |
+
"rewards/accuracies": 0.7437499761581421,
|
| 1224 |
+
"rewards/chosen": -6.083466053009033,
|
| 1225 |
+
"rewards/margins": 18.29672622680664,
|
| 1226 |
+
"rewards/rejected": -24.380191802978516,
|
| 1227 |
+
"step": 870
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
"epoch": 1.84,
|
| 1231 |
+
"learning_rate": 1.824189659787284e-08,
|
| 1232 |
+
"logits/chosen": -0.051941704005002975,
|
| 1233 |
+
"logits/rejected": -0.5002130270004272,
|
| 1234 |
+
"logps/chosen": -670.8041381835938,
|
| 1235 |
+
"logps/rejected": -2170.00439453125,
|
| 1236 |
+
"loss": 0.3627,
|
| 1237 |
"rewards/accuracies": 0.731249988079071,
|
| 1238 |
+
"rewards/chosen": -3.512953519821167,
|
| 1239 |
+
"rewards/margins": 15.10865306854248,
|
| 1240 |
+
"rewards/rejected": -18.621606826782227,
|
| 1241 |
+
"step": 880
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 1.86,
|
| 1245 |
+
"learning_rate": 1.3665854824458035e-08,
|
| 1246 |
+
"logits/chosen": -0.14072179794311523,
|
| 1247 |
+
"logits/rejected": -0.6090524792671204,
|
| 1248 |
+
"logps/chosen": -987.47119140625,
|
| 1249 |
+
"logps/rejected": -2067.42236328125,
|
| 1250 |
+
"loss": 0.3647,
|
| 1251 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1252 |
+
"rewards/chosen": -6.492043972015381,
|
| 1253 |
+
"rewards/margins": 11.107647895812988,
|
| 1254 |
+
"rewards/rejected": -17.599689483642578,
|
| 1255 |
+
"step": 890
|
| 1256 |
},
|
| 1257 |
{
|
| 1258 |
"epoch": 1.88,
|
| 1259 |
+
"learning_rate": 9.741758728888217e-09,
|
| 1260 |
+
"logits/chosen": -0.022175291553139687,
|
| 1261 |
+
"logits/rejected": -0.6860365271568298,
|
| 1262 |
+
"logps/chosen": -564.6074829101562,
|
| 1263 |
+
"logps/rejected": -2140.48681640625,
|
| 1264 |
+
"loss": 0.3545,
|
| 1265 |
+
"rewards/accuracies": 0.8187500238418579,
|
| 1266 |
+
"rewards/chosen": -2.2749757766723633,
|
| 1267 |
+
"rewards/margins": 15.992487907409668,
|
| 1268 |
+
"rewards/rejected": -18.2674617767334,
|
| 1269 |
+
"step": 900
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
"epoch": 1.91,
|
| 1273 |
+
"learning_rate": 6.474868681043577e-09,
|
| 1274 |
+
"logits/chosen": -0.09770497679710388,
|
| 1275 |
+
"logits/rejected": -0.611740231513977,
|
| 1276 |
+
"logps/chosen": -823.6361083984375,
|
| 1277 |
+
"logps/rejected": -2548.73388671875,
|
| 1278 |
+
"loss": 0.3561,
|
| 1279 |
+
"rewards/accuracies": 0.8062499761581421,
|
| 1280 |
+
"rewards/chosen": -5.1007561683654785,
|
| 1281 |
+
"rewards/margins": 17.229694366455078,
|
| 1282 |
+
"rewards/rejected": -22.330448150634766,
|
| 1283 |
+
"step": 910
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
"epoch": 1.93,
|
| 1287 |
+
"learning_rate": 3.869564046156459e-09,
|
| 1288 |
+
"logits/chosen": -0.051595211029052734,
|
| 1289 |
+
"logits/rejected": -0.6881163716316223,
|
| 1290 |
+
"logps/chosen": -818.4075927734375,
|
| 1291 |
+
"logps/rejected": -2222.89892578125,
|
| 1292 |
+
"loss": 0.3632,
|
| 1293 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 1294 |
+
"rewards/chosen": -5.061829566955566,
|
| 1295 |
+
"rewards/margins": 14.042346000671387,
|
| 1296 |
+
"rewards/rejected": -19.104175567626953,
|
| 1297 |
+
"step": 920
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
"epoch": 1.95,
|
| 1301 |
+
"learning_rate": 1.929337314139412e-09,
|
| 1302 |
+
"logits/chosen": -0.2066664695739746,
|
| 1303 |
+
"logits/rejected": -0.7445483803749084,
|
| 1304 |
+
"logps/chosen": -808.8455200195312,
|
| 1305 |
+
"logps/rejected": -2086.377197265625,
|
| 1306 |
+
"loss": 0.3643,
|
| 1307 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 1308 |
+
"rewards/chosen": -5.122731685638428,
|
| 1309 |
+
"rewards/margins": 12.991943359375,
|
| 1310 |
+
"rewards/rejected": -18.114675521850586,
|
| 1311 |
+
"step": 930
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
"epoch": 1.97,
|
| 1315 |
+
"learning_rate": 6.567894177967325e-10,
|
| 1316 |
+
"logits/chosen": 0.007280481047928333,
|
| 1317 |
+
"logits/rejected": -0.49907398223876953,
|
| 1318 |
+
"logps/chosen": -689.2498779296875,
|
| 1319 |
+
"logps/rejected": -2379.00048828125,
|
| 1320 |
+
"loss": 0.3578,
|
| 1321 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1322 |
+
"rewards/chosen": -3.5881881713867188,
|
| 1323 |
+
"rewards/margins": 17.178848266601562,
|
| 1324 |
+
"rewards/rejected": -20.767038345336914,
|
| 1325 |
+
"step": 940
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
"epoch": 1.99,
|
| 1329 |
+
"learning_rate": 5.3626246194704575e-11,
|
| 1330 |
+
"logits/chosen": -0.05890023708343506,
|
| 1331 |
+
"logits/rejected": -0.6455780863761902,
|
| 1332 |
+
"logps/chosen": -604.0885009765625,
|
| 1333 |
+
"logps/rejected": -2217.527099609375,
|
| 1334 |
+
"loss": 0.366,
|
| 1335 |
+
"rewards/accuracies": 0.8125,
|
| 1336 |
+
"rewards/chosen": -3.149109363555908,
|
| 1337 |
+
"rewards/margins": 16.183589935302734,
|
| 1338 |
+
"rewards/rejected": -19.332698822021484,
|
| 1339 |
+
"step": 950
|
| 1340 |
},
|
| 1341 |
{
|
| 1342 |
"epoch": 2.0,
|
| 1343 |
+
"step": 954,
|
| 1344 |
"total_flos": 0.0,
|
| 1345 |
+
"train_loss": 0.39703809490243847,
|
| 1346 |
+
"train_runtime": 12665.7954,
|
| 1347 |
+
"train_samples_per_second": 9.653,
|
| 1348 |
+
"train_steps_per_second": 0.075
|
| 1349 |
}
|
| 1350 |
],
|
| 1351 |
"logging_steps": 10,
|
| 1352 |
+
"max_steps": 954,
|
| 1353 |
"num_train_epochs": 2,
|
| 1354 |
"save_steps": 10000,
|
| 1355 |
"total_flos": 0.0,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6648
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1397e8512176db34249f7cc8cdf6db6939284d3b90a77a3bb793ba3757e0ba4c
|
| 3 |
size 6648
|