Model save
Browse files- README.md +2 -2
- all_results.json +4 -4
- 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 +4 -4
- trainer_state.json +863 -359
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -32,7 +32,7 @@ More information needed
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
-
- learning_rate:
|
| 36 |
- train_batch_size: 2
|
| 37 |
- eval_batch_size: 8
|
| 38 |
- seed: 1
|
|
@@ -44,7 +44,7 @@ The following hyperparameters were used during training:
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: cosine
|
| 46 |
- lr_scheduler_warmup_ratio: 0.1
|
| 47 |
-
- num_epochs:
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
|
|
|
| 32 |
### Training hyperparameters
|
| 33 |
|
| 34 |
The following hyperparameters were used during training:
|
| 35 |
+
- learning_rate: 2e-06
|
| 36 |
- train_batch_size: 2
|
| 37 |
- eval_batch_size: 8
|
| 38 |
- seed: 1
|
|
|
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: cosine
|
| 46 |
- lr_scheduler_warmup_ratio: 0.1
|
| 47 |
+
- num_epochs: 2
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 45548,
|
| 6 |
-
"train_samples_per_second": 11.
|
| 7 |
"train_steps_per_second": 0.087
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.18113358499298632,
|
| 4 |
+
"train_runtime": 8124.3794,
|
| 5 |
"train_samples": 45548,
|
| 6 |
+
"train_samples_per_second": 11.213,
|
| 7 |
"train_steps_per_second": 0.087
|
| 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 sha256:5e0f852f610b5179bc0d46ffe89d5644556da8841082b008cdae3c65d2a759bb
|
| 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:c2b594dde7a8f834fedff85680da6838d81c3edbfb2b21c505fa6c3c19bf968f
|
| 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:efd9bdaaa5dc9e6569743b905fb69b653c0deaca599081e9a0b714570bd74b7b
|
| 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 sha256:4291b6d92a0c55352fc37c06c9f74edafeb0cfa5b292a4fa380c79ded3be2a05
|
| 3 |
size 1168138808
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 45548,
|
| 6 |
-
"train_samples_per_second": 11.
|
| 7 |
"train_steps_per_second": 0.087
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.18113358499298632,
|
| 4 |
+
"train_runtime": 8124.3794,
|
| 5 |
"train_samples": 45548,
|
| 6 |
+
"train_samples_per_second": 11.213,
|
| 7 |
"train_steps_per_second": 0.087
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,16 +1,16 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 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": 0.0,
|
| 13 |
-
"learning_rate":
|
| 14 |
"logits/chosen": -0.20257116854190826,
|
| 15 |
"logits/rejected": -0.07568661123514175,
|
| 16 |
"logps/chosen": -521.0269775390625,
|
|
@@ -24,507 +24,1011 @@
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03,
|
| 27 |
-
"learning_rate":
|
| 28 |
-
"logits/chosen": -0.
|
| 29 |
-
"logits/rejected": -0.
|
| 30 |
-
"logps/chosen": -345.
|
| 31 |
-
"logps/rejected": -280.
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"rewards/accuracies": 0.
|
| 34 |
-
"rewards/chosen":
|
| 35 |
-
"rewards/margins": -0.
|
| 36 |
-
"rewards/rejected": 0.
|
| 37 |
"step": 10
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.06,
|
| 41 |
-
"learning_rate":
|
| 42 |
-
"logits/chosen": -0.
|
| 43 |
-
"logits/rejected": -0.
|
| 44 |
-
"logps/chosen": -331.
|
| 45 |
-
"logps/rejected": -241.
|
| 46 |
-
"loss": 0.
|
| 47 |
-
"rewards/accuracies": 0.
|
| 48 |
-
"rewards/chosen": 0.
|
| 49 |
-
"rewards/margins": 0.
|
| 50 |
-
"rewards/rejected": -0.
|
| 51 |
"step": 20
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.08,
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"logits/chosen": -0.
|
| 57 |
-
"logits/rejected":
|
| 58 |
-
"logps/chosen": -387.
|
| 59 |
-
"logps/rejected": -
|
| 60 |
-
"loss": 0.
|
| 61 |
-
"rewards/accuracies": 0.
|
| 62 |
-
"rewards/chosen": 0.
|
| 63 |
-
"rewards/margins": 0.
|
| 64 |
-
"rewards/rejected": -0.
|
| 65 |
"step": 30
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.11,
|
| 69 |
-
"learning_rate":
|
| 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": -0.
|
| 79 |
"step": 40
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.14,
|
| 83 |
-
"learning_rate":
|
| 84 |
-
"logits/chosen": -0.
|
| 85 |
-
"logits/rejected": 0.
|
| 86 |
-
"logps/chosen": -
|
| 87 |
-
"logps/rejected": -
|
| 88 |
-
"loss": 0.
|
| 89 |
-
"rewards/accuracies": 0.
|
| 90 |
-
"rewards/chosen": -0.
|
| 91 |
-
"rewards/margins": 0.
|
| 92 |
-
"rewards/rejected": -0.
|
| 93 |
"step": 50
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.17,
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"logits/chosen": -0.
|
| 99 |
-
"logits/rejected": -0.
|
| 100 |
-
"logps/chosen": -
|
| 101 |
-
"logps/rejected": -
|
| 102 |
-
"loss": 0.
|
| 103 |
-
"rewards/accuracies": 0.
|
| 104 |
-
"rewards/chosen": -0.
|
| 105 |
-
"rewards/margins": 0.
|
| 106 |
-
"rewards/rejected": -0.
|
| 107 |
"step": 60
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.2,
|
| 111 |
-
"learning_rate":
|
| 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": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.22,
|
| 125 |
-
"learning_rate":
|
| 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": 80
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.25,
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"logits/chosen": -0.
|
| 141 |
-
"logits/rejected": -0.
|
| 142 |
-
"logps/chosen": -
|
| 143 |
-
"logps/rejected": -
|
| 144 |
-
"loss": 0.
|
| 145 |
-
"rewards/accuracies": 0.
|
| 146 |
-
"rewards/chosen": 0.
|
| 147 |
-
"rewards/margins": 0.
|
| 148 |
-
"rewards/rejected": -0.
|
| 149 |
"step": 90
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.28,
|
| 153 |
-
"learning_rate":
|
| 154 |
-
"logits/chosen": -0.
|
| 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": 100
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.31,
|
| 167 |
-
"learning_rate":
|
| 168 |
-
"logits/chosen": -0.
|
| 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": 110
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.34,
|
| 181 |
-
"learning_rate":
|
| 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": -0.
|
| 191 |
"step": 120
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.37,
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"logits/chosen": -0.
|
| 197 |
-
"logits/rejected": -0.
|
| 198 |
-
"logps/chosen": -
|
| 199 |
-
"logps/rejected": -
|
| 200 |
-
"loss": 0.
|
| 201 |
-
"rewards/accuracies": 0.
|
| 202 |
-
"rewards/chosen": 0.
|
| 203 |
-
"rewards/margins": 0.
|
| 204 |
-
"rewards/rejected": -0.
|
| 205 |
"step": 130
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.39,
|
| 209 |
-
"learning_rate":
|
| 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": 0.
|
| 218 |
-
"rewards/rejected": -0.
|
| 219 |
"step": 140
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.42,
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"logits/chosen": -0.
|
| 225 |
-
"logits/rejected":
|
| 226 |
-
"logps/chosen": -
|
| 227 |
-
"logps/rejected": -
|
| 228 |
-
"loss": 0.
|
| 229 |
-
"rewards/accuracies": 0.
|
| 230 |
-
"rewards/chosen": 0.
|
| 231 |
-
"rewards/margins": 0.
|
| 232 |
-
"rewards/rejected": -0.
|
| 233 |
"step": 150
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.45,
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"logits/chosen": -0.
|
| 239 |
-
"logits/rejected":
|
| 240 |
-
"logps/chosen": -
|
| 241 |
-
"logps/rejected": -
|
| 242 |
-
"loss": 0.
|
| 243 |
-
"rewards/accuracies": 0.
|
| 244 |
-
"rewards/chosen": 0.
|
| 245 |
-
"rewards/margins": 0.
|
| 246 |
-
"rewards/rejected": -0.
|
| 247 |
"step": 160
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.48,
|
| 251 |
-
"learning_rate":
|
| 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": 0.
|
| 260 |
-
"rewards/rejected": -0.
|
| 261 |
"step": 170
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.51,
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"logits/chosen": -0.
|
| 267 |
-
"logits/rejected": -0.
|
| 268 |
-
"logps/chosen": -
|
| 269 |
-
"logps/rejected": -
|
| 270 |
-
"loss": 0.
|
| 271 |
-
"rewards/accuracies": 0.
|
| 272 |
-
"rewards/chosen": -0.
|
| 273 |
-
"rewards/margins": 0.
|
| 274 |
-
"rewards/rejected": -0.
|
| 275 |
"step": 180
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.53,
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"logits/chosen": -0.
|
| 281 |
-
"logits/rejected":
|
| 282 |
-
"logps/chosen": -
|
| 283 |
-
"logps/rejected": -
|
| 284 |
-
"loss": 0.
|
| 285 |
-
"rewards/accuracies": 0.
|
| 286 |
-
"rewards/chosen": 0.
|
| 287 |
-
"rewards/margins": 0.
|
| 288 |
-
"rewards/rejected": -0.
|
| 289 |
"step": 190
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.56,
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"logits/chosen": -0.
|
| 295 |
-
"logits/rejected":
|
| 296 |
-
"logps/chosen": -
|
| 297 |
-
"logps/rejected": -
|
| 298 |
-
"loss": 0.
|
| 299 |
-
"rewards/accuracies": 0.
|
| 300 |
-
"rewards/chosen": 0.
|
| 301 |
-
"rewards/margins": 0.
|
| 302 |
-
"rewards/rejected": -0.
|
| 303 |
"step": 200
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.59,
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"logits/chosen":
|
| 309 |
-
"logits/rejected":
|
| 310 |
-
"logps/chosen": -
|
| 311 |
-
"logps/rejected": -
|
| 312 |
-
"loss": 0.
|
| 313 |
-
"rewards/accuracies": 0.
|
| 314 |
-
"rewards/chosen": 0.
|
| 315 |
-
"rewards/margins": 0.
|
| 316 |
-
"rewards/rejected": -0.
|
| 317 |
"step": 210
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.62,
|
| 321 |
-
"learning_rate":
|
| 322 |
-
"logits/chosen":
|
| 323 |
-
"logits/rejected":
|
| 324 |
-
"logps/chosen": -
|
| 325 |
-
"logps/rejected": -
|
| 326 |
-
"loss": 0.
|
| 327 |
-
"rewards/accuracies": 0.
|
| 328 |
-
"rewards/chosen": -0.
|
| 329 |
-
"rewards/margins": 0.
|
| 330 |
-
"rewards/rejected": -0.
|
| 331 |
"step": 220
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.65,
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"logits/chosen":
|
| 337 |
-
"logits/rejected": 0.
|
| 338 |
-
"logps/chosen": -
|
| 339 |
-
"logps/rejected": -
|
| 340 |
-
"loss": 0.
|
| 341 |
-
"rewards/accuracies": 0.
|
| 342 |
-
"rewards/chosen": 0.
|
| 343 |
-
"rewards/margins": 0.
|
| 344 |
-
"rewards/rejected": -0.
|
| 345 |
"step": 230
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.67,
|
| 349 |
-
"learning_rate": 1.
|
| 350 |
-
"logits/chosen": -0.
|
| 351 |
-
"logits/rejected": -0.
|
| 352 |
-
"logps/chosen": -
|
| 353 |
-
"logps/rejected": -
|
| 354 |
-
"loss": 0.
|
| 355 |
-
"rewards/accuracies": 0.
|
| 356 |
-
"rewards/chosen": 0.
|
| 357 |
-
"rewards/margins": 0.
|
| 358 |
-
"rewards/rejected": -0.
|
| 359 |
"step": 240
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.7,
|
| 363 |
-
"learning_rate": 1.
|
| 364 |
-
"logits/chosen": -0.
|
| 365 |
-
"logits/rejected": -0.
|
| 366 |
-
"logps/chosen": -
|
| 367 |
-
"logps/rejected": -
|
| 368 |
-
"loss": 0.
|
| 369 |
-
"rewards/accuracies": 0.
|
| 370 |
-
"rewards/chosen": 0.
|
| 371 |
-
"rewards/margins": 0.
|
| 372 |
-
"rewards/rejected": -0.
|
| 373 |
"step": 250
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.73,
|
| 377 |
-
"learning_rate": 1.
|
| 378 |
-
"logits/chosen": -0.
|
| 379 |
-
"logits/rejected":
|
| 380 |
-
"logps/chosen": -
|
| 381 |
-
"logps/rejected": -
|
| 382 |
-
"loss": 0.
|
| 383 |
-
"rewards/accuracies": 0.
|
| 384 |
-
"rewards/chosen": 0.
|
| 385 |
-
"rewards/margins": 0.
|
| 386 |
-
"rewards/rejected": -
|
| 387 |
"step": 260
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.76,
|
| 391 |
-
"learning_rate":
|
| 392 |
-
"logits/chosen":
|
| 393 |
-
"logits/rejected":
|
| 394 |
-
"logps/chosen": -
|
| 395 |
-
"logps/rejected": -
|
| 396 |
-
"loss": 0.
|
| 397 |
"rewards/accuracies": 0.606249988079071,
|
| 398 |
-
"rewards/chosen": -0.
|
| 399 |
-
"rewards/margins": 0.
|
| 400 |
-
"rewards/rejected": -
|
| 401 |
"step": 270
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.79,
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"logits/chosen":
|
| 407 |
-
"logits/rejected": 0.
|
| 408 |
-
"logps/chosen": -
|
| 409 |
-
"logps/rejected": -
|
| 410 |
-
"loss": 0.
|
| 411 |
-
"rewards/accuracies": 0.
|
| 412 |
-
"rewards/chosen": 0.
|
| 413 |
-
"rewards/margins": 0.
|
| 414 |
-
"rewards/rejected": -
|
| 415 |
"step": 280
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 0.81,
|
| 419 |
-
"learning_rate":
|
| 420 |
-
"logits/chosen":
|
| 421 |
-
"logits/rejected": 0.
|
| 422 |
-
"logps/chosen": -
|
| 423 |
-
"logps/rejected": -
|
| 424 |
-
"loss": 0.
|
| 425 |
-
"rewards/accuracies": 0.
|
| 426 |
-
"rewards/chosen": -0.
|
| 427 |
-
"rewards/margins": 0.
|
| 428 |
-
"rewards/rejected": -
|
| 429 |
"step": 290
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.84,
|
| 433 |
-
"learning_rate":
|
| 434 |
-
"logits/chosen": -0.
|
| 435 |
-
"logits/rejected": 0.
|
| 436 |
-
"logps/chosen": -
|
| 437 |
-
"logps/rejected": -
|
| 438 |
-
"loss": 0.
|
| 439 |
-
"rewards/accuracies": 0.
|
| 440 |
-
"rewards/chosen": 0.
|
| 441 |
-
"rewards/margins": 0.
|
| 442 |
-
"rewards/rejected": -0.
|
| 443 |
"step": 300
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 0.87,
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"logits/chosen": -0.
|
| 449 |
-
"logits/rejected":
|
| 450 |
-
"logps/chosen": -
|
| 451 |
-
"logps/rejected": -
|
| 452 |
-
"loss": 0.
|
| 453 |
-
"rewards/accuracies": 0.
|
| 454 |
-
"rewards/chosen": -0.
|
| 455 |
-
"rewards/margins": 0.
|
| 456 |
-
"rewards/rejected": -0.
|
| 457 |
"step": 310
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.9,
|
| 461 |
-
"learning_rate": 1.
|
| 462 |
-
"logits/chosen":
|
| 463 |
-
"logits/rejected": 0.
|
| 464 |
-
"logps/chosen": -
|
| 465 |
-
"logps/rejected": -
|
| 466 |
-
"loss": 0.
|
| 467 |
-
"rewards/accuracies": 0.
|
| 468 |
-
"rewards/chosen": -0.
|
| 469 |
-
"rewards/margins": 0.
|
| 470 |
-
"rewards/rejected": -
|
| 471 |
"step": 320
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 0.93,
|
| 475 |
-
"learning_rate":
|
| 476 |
-
"logits/chosen": -0.
|
| 477 |
-
"logits/rejected":
|
| 478 |
-
"logps/chosen": -
|
| 479 |
-
"logps/rejected": -
|
| 480 |
-
"loss": 0.
|
| 481 |
-
"rewards/accuracies": 0.
|
| 482 |
-
"rewards/chosen": 0.
|
| 483 |
-
"rewards/margins": 0.
|
| 484 |
-
"rewards/rejected": -
|
| 485 |
"step": 330
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"epoch": 0.96,
|
| 489 |
-
"learning_rate":
|
| 490 |
-
"logits/chosen": -0.
|
| 491 |
-
"logits/rejected":
|
| 492 |
-
"logps/chosen": -
|
| 493 |
-
"logps/rejected": -
|
| 494 |
-
"loss": 0.
|
| 495 |
-
"rewards/accuracies": 0.
|
| 496 |
-
"rewards/chosen": -0.
|
| 497 |
-
"rewards/margins": 0.
|
| 498 |
-
"rewards/rejected": -
|
| 499 |
"step": 340
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 0.98,
|
| 503 |
-
"learning_rate":
|
| 504 |
-
"logits/chosen": -0.
|
| 505 |
-
"logits/rejected":
|
| 506 |
-
"logps/chosen": -
|
| 507 |
-
"logps/rejected": -
|
| 508 |
-
"loss": 0.
|
| 509 |
-
"rewards/accuracies": 0.
|
| 510 |
-
"rewards/chosen": -0.
|
| 511 |
-
"rewards/margins": 0.
|
| 512 |
-
"rewards/rejected": -
|
| 513 |
"step": 350
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 1.
|
| 517 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 518 |
"total_flos": 0.0,
|
| 519 |
-
"train_loss": 0.
|
| 520 |
-
"train_runtime":
|
| 521 |
-
"train_samples_per_second": 11.
|
| 522 |
"train_steps_per_second": 0.087
|
| 523 |
}
|
| 524 |
],
|
| 525 |
"logging_steps": 10,
|
| 526 |
-
"max_steps":
|
| 527 |
-
"num_train_epochs":
|
| 528 |
"save_steps": 10000,
|
| 529 |
"total_flos": 0.0,
|
| 530 |
"trial_name": null,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9950825430277486,
|
| 5 |
"eval_steps": 10000,
|
| 6 |
+
"global_step": 710,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 2.816901408450704e-08,
|
| 14 |
"logits/chosen": -0.20257116854190826,
|
| 15 |
"logits/rejected": -0.07568661123514175,
|
| 16 |
"logps/chosen": -521.0269775390625,
|
|
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03,
|
| 27 |
+
"learning_rate": 2.8169014084507043e-07,
|
| 28 |
+
"logits/chosen": -0.13755783438682556,
|
| 29 |
+
"logits/rejected": -0.08772192895412445,
|
| 30 |
+
"logps/chosen": -345.8690185546875,
|
| 31 |
+
"logps/rejected": -280.64691162109375,
|
| 32 |
+
"loss": 0.3323,
|
| 33 |
+
"rewards/accuracies": 0.4513888955116272,
|
| 34 |
+
"rewards/chosen": 4.912400618195534e-05,
|
| 35 |
+
"rewards/margins": -0.00030259901541285217,
|
| 36 |
+
"rewards/rejected": 0.0003517230798024684,
|
| 37 |
"step": 10
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.06,
|
| 41 |
+
"learning_rate": 5.633802816901409e-07,
|
| 42 |
+
"logits/chosen": -0.0957806184887886,
|
| 43 |
+
"logits/rejected": -0.057603687047958374,
|
| 44 |
+
"logps/chosen": -331.79766845703125,
|
| 45 |
+
"logps/rejected": -241.81185913085938,
|
| 46 |
+
"loss": 0.329,
|
| 47 |
+
"rewards/accuracies": 0.550000011920929,
|
| 48 |
+
"rewards/chosen": 0.0004024081281386316,
|
| 49 |
+
"rewards/margins": 0.00799016747623682,
|
| 50 |
+
"rewards/rejected": -0.007587759289890528,
|
| 51 |
"step": 20
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.08,
|
| 55 |
+
"learning_rate": 8.450704225352112e-07,
|
| 56 |
+
"logits/chosen": -0.05522637814283371,
|
| 57 |
+
"logits/rejected": 0.0007691409555263817,
|
| 58 |
+
"logps/chosen": -387.5187072753906,
|
| 59 |
+
"logps/rejected": -256.33380126953125,
|
| 60 |
+
"loss": 0.3393,
|
| 61 |
+
"rewards/accuracies": 0.606249988079071,
|
| 62 |
+
"rewards/chosen": 0.0015848552575334907,
|
| 63 |
+
"rewards/margins": 0.056810565292835236,
|
| 64 |
+
"rewards/rejected": -0.05522570759057999,
|
| 65 |
"step": 30
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.11,
|
| 69 |
+
"learning_rate": 1.1267605633802817e-06,
|
| 70 |
+
"logits/chosen": -0.07449092715978622,
|
| 71 |
+
"logits/rejected": -0.021578723564743996,
|
| 72 |
+
"logps/chosen": -341.57391357421875,
|
| 73 |
+
"logps/rejected": -272.60400390625,
|
| 74 |
+
"loss": 0.3699,
|
| 75 |
+
"rewards/accuracies": 0.518750011920929,
|
| 76 |
+
"rewards/chosen": -0.0639580637216568,
|
| 77 |
+
"rewards/margins": 0.056426145136356354,
|
| 78 |
+
"rewards/rejected": -0.12038421630859375,
|
| 79 |
"step": 40
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.14,
|
| 83 |
+
"learning_rate": 1.408450704225352e-06,
|
| 84 |
+
"logits/chosen": -0.06206550449132919,
|
| 85 |
+
"logits/rejected": -0.011999653652310371,
|
| 86 |
+
"logps/chosen": -344.6331787109375,
|
| 87 |
+
"logps/rejected": -287.69635009765625,
|
| 88 |
+
"loss": 0.3883,
|
| 89 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 90 |
+
"rewards/chosen": -0.10917987674474716,
|
| 91 |
+
"rewards/margins": 0.14197741448879242,
|
| 92 |
+
"rewards/rejected": -0.251157283782959,
|
| 93 |
"step": 50
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.17,
|
| 97 |
+
"learning_rate": 1.6901408450704225e-06,
|
| 98 |
+
"logits/chosen": -0.17808644473552704,
|
| 99 |
+
"logits/rejected": -0.11181376129388809,
|
| 100 |
+
"logps/chosen": -425.6441345214844,
|
| 101 |
+
"logps/rejected": -281.8149108886719,
|
| 102 |
+
"loss": 0.3986,
|
| 103 |
+
"rewards/accuracies": 0.6875,
|
| 104 |
+
"rewards/chosen": -0.03321179002523422,
|
| 105 |
+
"rewards/margins": 0.33166056871414185,
|
| 106 |
+
"rewards/rejected": -0.36487236618995667,
|
| 107 |
"step": 60
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.2,
|
| 111 |
+
"learning_rate": 1.971830985915493e-06,
|
| 112 |
+
"logits/chosen": -0.17757034301757812,
|
| 113 |
+
"logits/rejected": -0.054698117077350616,
|
| 114 |
+
"logps/chosen": -403.33819580078125,
|
| 115 |
+
"logps/rejected": -279.6980895996094,
|
| 116 |
+
"loss": 0.3837,
|
| 117 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 118 |
+
"rewards/chosen": 0.03798174858093262,
|
| 119 |
+
"rewards/margins": 0.2265440672636032,
|
| 120 |
+
"rewards/rejected": -0.1885623335838318,
|
| 121 |
"step": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.22,
|
| 125 |
+
"learning_rate": 1.9990212265199736e-06,
|
| 126 |
+
"logits/chosen": -0.1993887722492218,
|
| 127 |
+
"logits/rejected": -0.19502341747283936,
|
| 128 |
+
"logps/chosen": -284.59423828125,
|
| 129 |
+
"logps/rejected": -219.568359375,
|
| 130 |
+
"loss": 0.3918,
|
| 131 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 132 |
+
"rewards/chosen": 0.027263695374131203,
|
| 133 |
+
"rewards/margins": 0.18515612185001373,
|
| 134 |
+
"rewards/rejected": -0.1578924059867859,
|
| 135 |
"step": 80
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.25,
|
| 139 |
+
"learning_rate": 1.995640271796129e-06,
|
| 140 |
+
"logits/chosen": -0.3508986830711365,
|
| 141 |
+
"logits/rejected": -0.2989083230495453,
|
| 142 |
+
"logps/chosen": -388.96124267578125,
|
| 143 |
+
"logps/rejected": -249.9578094482422,
|
| 144 |
+
"loss": 0.3972,
|
| 145 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 146 |
+
"rewards/chosen": 0.1189202070236206,
|
| 147 |
+
"rewards/margins": 0.3718037009239197,
|
| 148 |
+
"rewards/rejected": -0.25288352370262146,
|
| 149 |
"step": 90
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.28,
|
| 153 |
+
"learning_rate": 1.9898532207817787e-06,
|
| 154 |
+
"logits/chosen": -0.13028475642204285,
|
| 155 |
+
"logits/rejected": -0.12288811057806015,
|
| 156 |
+
"logps/chosen": -375.6239929199219,
|
| 157 |
+
"logps/rejected": -300.58013916015625,
|
| 158 |
+
"loss": 0.3681,
|
| 159 |
+
"rewards/accuracies": 0.612500011920929,
|
| 160 |
+
"rewards/chosen": -0.078019879758358,
|
| 161 |
+
"rewards/margins": 0.1508217304944992,
|
| 162 |
+
"rewards/rejected": -0.2288416177034378,
|
| 163 |
"step": 100
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.31,
|
| 167 |
+
"learning_rate": 1.9816740586504575e-06,
|
| 168 |
+
"logits/chosen": -0.09742162376642227,
|
| 169 |
+
"logits/rejected": -0.04224073141813278,
|
| 170 |
+
"logps/chosen": -351.0619201660156,
|
| 171 |
+
"logps/rejected": -310.95257568359375,
|
| 172 |
+
"loss": 0.3116,
|
| 173 |
+
"rewards/accuracies": 0.53125,
|
| 174 |
+
"rewards/chosen": -0.23143887519836426,
|
| 175 |
+
"rewards/margins": 0.0850377082824707,
|
| 176 |
+
"rewards/rejected": -0.3164765536785126,
|
| 177 |
"step": 110
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.34,
|
| 181 |
+
"learning_rate": 1.971122551428331e-06,
|
| 182 |
+
"logits/chosen": -0.10740338265895844,
|
| 183 |
+
"logits/rejected": -0.07498879730701447,
|
| 184 |
+
"logps/chosen": -354.75677490234375,
|
| 185 |
+
"logps/rejected": -261.3316955566406,
|
| 186 |
+
"loss": 0.3164,
|
| 187 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 188 |
+
"rewards/chosen": -0.22371773421764374,
|
| 189 |
+
"rewards/margins": 0.21556267142295837,
|
| 190 |
+
"rewards/rejected": -0.4392804205417633,
|
| 191 |
"step": 120
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.37,
|
| 195 |
+
"learning_rate": 1.9582241982269803e-06,
|
| 196 |
+
"logits/chosen": -0.090018130838871,
|
| 197 |
+
"logits/rejected": -0.013507525436580181,
|
| 198 |
+
"logps/chosen": -415.02020263671875,
|
| 199 |
+
"logps/rejected": -290.92144775390625,
|
| 200 |
+
"loss": 0.3473,
|
| 201 |
+
"rewards/accuracies": 0.6875,
|
| 202 |
+
"rewards/chosen": -0.28872472047805786,
|
| 203 |
+
"rewards/margins": 0.3263753056526184,
|
| 204 |
+
"rewards/rejected": -0.6151000261306763,
|
| 205 |
"step": 130
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.39,
|
| 209 |
+
"learning_rate": 1.9430101696214336e-06,
|
| 210 |
+
"logits/chosen": -0.1004706472158432,
|
| 211 |
+
"logits/rejected": -0.08849872648715973,
|
| 212 |
+
"logps/chosen": -363.7771301269531,
|
| 213 |
+
"logps/rejected": -290.38299560546875,
|
| 214 |
+
"loss": 0.3338,
|
| 215 |
+
"rewards/accuracies": 0.581250011920929,
|
| 216 |
+
"rewards/chosen": -0.2705627381801605,
|
| 217 |
+
"rewards/margins": 0.14047202467918396,
|
| 218 |
+
"rewards/rejected": -0.4110347628593445,
|
| 219 |
"step": 140
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.42,
|
| 223 |
+
"learning_rate": 1.9255172323223463e-06,
|
| 224 |
+
"logits/chosen": -0.04992828518152237,
|
| 225 |
+
"logits/rejected": 0.011964783072471619,
|
| 226 |
+
"logps/chosen": -336.70928955078125,
|
| 227 |
+
"logps/rejected": -298.07611083984375,
|
| 228 |
+
"loss": 0.3193,
|
| 229 |
+
"rewards/accuracies": 0.5562499761581421,
|
| 230 |
+
"rewards/chosen": -0.31615492701530457,
|
| 231 |
+
"rewards/margins": 0.1844826191663742,
|
| 232 |
+
"rewards/rejected": -0.5006375312805176,
|
| 233 |
"step": 150
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.45,
|
| 237 |
+
"learning_rate": 1.905787660324391e-06,
|
| 238 |
+
"logits/chosen": -0.011503048241138458,
|
| 239 |
+
"logits/rejected": 0.1220928207039833,
|
| 240 |
+
"logps/chosen": -424.7059631347656,
|
| 241 |
+
"logps/rejected": -320.0885009765625,
|
| 242 |
+
"loss": 0.3057,
|
| 243 |
+
"rewards/accuracies": 0.612500011920929,
|
| 244 |
+
"rewards/chosen": -0.3887358605861664,
|
| 245 |
+
"rewards/margins": 0.23610559105873108,
|
| 246 |
+
"rewards/rejected": -0.6248414516448975,
|
| 247 |
"step": 160
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.48,
|
| 251 |
+
"learning_rate": 1.8838691327455609e-06,
|
| 252 |
+
"logits/chosen": -0.026822054758667946,
|
| 253 |
+
"logits/rejected": 0.04267163202166557,
|
| 254 |
+
"logps/chosen": -369.6314392089844,
|
| 255 |
+
"logps/rejected": -326.5650329589844,
|
| 256 |
+
"loss": 0.3049,
|
| 257 |
+
"rewards/accuracies": 0.612500011920929,
|
| 258 |
+
"rewards/chosen": -0.37862712144851685,
|
| 259 |
+
"rewards/margins": 0.2466701716184616,
|
| 260 |
+
"rewards/rejected": -0.6252972483634949,
|
| 261 |
"step": 170
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.51,
|
| 265 |
+
"learning_rate": 1.8598146186042808e-06,
|
| 266 |
+
"logits/chosen": -0.2679498791694641,
|
| 267 |
+
"logits/rejected": -0.21142823994159698,
|
| 268 |
+
"logps/chosen": -383.9067687988281,
|
| 269 |
+
"logps/rejected": -319.5235900878906,
|
| 270 |
+
"loss": 0.3293,
|
| 271 |
+
"rewards/accuracies": 0.612500011920929,
|
| 272 |
+
"rewards/chosen": -0.4264276623725891,
|
| 273 |
+
"rewards/margins": 0.24799779057502747,
|
| 274 |
+
"rewards/rejected": -0.674425482749939,
|
| 275 |
"step": 180
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.53,
|
| 279 |
+
"learning_rate": 1.8336822488127723e-06,
|
| 280 |
+
"logits/chosen": -0.05206901952624321,
|
| 281 |
+
"logits/rejected": 0.03507734090089798,
|
| 282 |
+
"logps/chosen": -408.74798583984375,
|
| 283 |
+
"logps/rejected": -323.0229797363281,
|
| 284 |
+
"loss": 0.284,
|
| 285 |
+
"rewards/accuracies": 0.643750011920929,
|
| 286 |
+
"rewards/chosen": -0.5230687260627747,
|
| 287 |
+
"rewards/margins": 0.4133565425872803,
|
| 288 |
+
"rewards/rejected": -0.9364253282546997,
|
| 289 |
"step": 190
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.56,
|
| 293 |
+
"learning_rate": 1.805535175696026e-06,
|
| 294 |
+
"logits/chosen": -0.026757067069411278,
|
| 295 |
+
"logits/rejected": 0.11062588542699814,
|
| 296 |
+
"logps/chosen": -406.74176025390625,
|
| 297 |
+
"logps/rejected": -343.10546875,
|
| 298 |
+
"loss": 0.2943,
|
| 299 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 300 |
+
"rewards/chosen": -0.7670020461082458,
|
| 301 |
+
"rewards/margins": 0.22102268040180206,
|
| 302 |
+
"rewards/rejected": -0.9880247116088867,
|
| 303 |
"step": 200
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.59,
|
| 307 |
+
"learning_rate": 1.7754414203758602e-06,
|
| 308 |
+
"logits/chosen": 0.09901715815067291,
|
| 309 |
+
"logits/rejected": 0.19180616736412048,
|
| 310 |
+
"logps/chosen": -463.052734375,
|
| 311 |
+
"logps/rejected": -346.2613220214844,
|
| 312 |
+
"loss": 0.2641,
|
| 313 |
+
"rewards/accuracies": 0.6312500238418579,
|
| 314 |
+
"rewards/chosen": -0.6034746766090393,
|
| 315 |
+
"rewards/margins": 0.1767103672027588,
|
| 316 |
+
"rewards/rejected": -0.7801850438117981,
|
| 317 |
"step": 210
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.62,
|
| 321 |
+
"learning_rate": 1.7434737083888904e-06,
|
| 322 |
+
"logits/chosen": 0.036244794726371765,
|
| 323 |
+
"logits/rejected": 0.10729759931564331,
|
| 324 |
+
"logps/chosen": -457.8307189941406,
|
| 325 |
+
"logps/rejected": -356.2217712402344,
|
| 326 |
+
"loss": 0.2726,
|
| 327 |
+
"rewards/accuracies": 0.6812499761581421,
|
| 328 |
+
"rewards/chosen": -0.6712073087692261,
|
| 329 |
+
"rewards/margins": 0.2408912628889084,
|
| 330 |
+
"rewards/rejected": -0.9120985865592957,
|
| 331 |
"step": 220
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.65,
|
| 335 |
+
"learning_rate": 1.709709293935662e-06,
|
| 336 |
+
"logits/chosen": 0.09133528172969818,
|
| 337 |
+
"logits/rejected": 0.2221805602312088,
|
| 338 |
+
"logps/chosen": -408.2891540527344,
|
| 339 |
+
"logps/rejected": -331.88568115234375,
|
| 340 |
+
"loss": 0.2977,
|
| 341 |
+
"rewards/accuracies": 0.612500011920929,
|
| 342 |
+
"rewards/chosen": -0.5791844129562378,
|
| 343 |
+
"rewards/margins": 0.30291682481765747,
|
| 344 |
+
"rewards/rejected": -0.8821012377738953,
|
| 345 |
"step": 230
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.67,
|
| 349 |
+
"learning_rate": 1.6742297731856636e-06,
|
| 350 |
+
"logits/chosen": -0.13875767588615417,
|
| 351 |
+
"logits/rejected": -0.030141670256853104,
|
| 352 |
+
"logps/chosen": -401.0727233886719,
|
| 353 |
+
"logps/rejected": -305.03875732421875,
|
| 354 |
+
"loss": 0.301,
|
| 355 |
+
"rewards/accuracies": 0.65625,
|
| 356 |
+
"rewards/chosen": -0.4225517213344574,
|
| 357 |
+
"rewards/margins": 0.2361893206834793,
|
| 358 |
+
"rewards/rejected": -0.6587410569190979,
|
| 359 |
"step": 240
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.7,
|
| 363 |
+
"learning_rate": 1.6371208870894001e-06,
|
| 364 |
+
"logits/chosen": -0.11314062029123306,
|
| 365 |
+
"logits/rejected": -0.028332043439149857,
|
| 366 |
+
"logps/chosen": -430.2290954589844,
|
| 367 |
+
"logps/rejected": -321.7243957519531,
|
| 368 |
+
"loss": 0.3146,
|
| 369 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 370 |
+
"rewards/chosen": -0.48336172103881836,
|
| 371 |
+
"rewards/margins": 0.32641273736953735,
|
| 372 |
+
"rewards/rejected": -0.8097743988037109,
|
| 373 |
"step": 250
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.73,
|
| 377 |
+
"learning_rate": 1.5984723141740574e-06,
|
| 378 |
+
"logits/chosen": -0.060978569090366364,
|
| 379 |
+
"logits/rejected": 0.013845810666680336,
|
| 380 |
+
"logps/chosen": -429.4888610839844,
|
| 381 |
+
"logps/rejected": -350.95147705078125,
|
| 382 |
+
"loss": 0.3233,
|
| 383 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 384 |
+
"rewards/chosen": -0.6816853880882263,
|
| 385 |
+
"rewards/margins": 0.37134552001953125,
|
| 386 |
+
"rewards/rejected": -1.0530308485031128,
|
| 387 |
"step": 260
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.76,
|
| 391 |
+
"learning_rate": 1.5583774538234882e-06,
|
| 392 |
+
"logits/chosen": 0.009507184848189354,
|
| 393 |
+
"logits/rejected": 0.053323499858379364,
|
| 394 |
+
"logps/chosen": -420.3421325683594,
|
| 395 |
+
"logps/rejected": -358.8752136230469,
|
| 396 |
+
"loss": 0.2801,
|
| 397 |
"rewards/accuracies": 0.606249988079071,
|
| 398 |
+
"rewards/chosen": -0.8152508735656738,
|
| 399 |
+
"rewards/margins": 0.3313834071159363,
|
| 400 |
+
"rewards/rejected": -1.1466342210769653,
|
| 401 |
"step": 270
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.79,
|
| 405 |
+
"learning_rate": 1.5169332005662589e-06,
|
| 406 |
+
"logits/chosen": 0.036943964660167694,
|
| 407 |
+
"logits/rejected": 0.12250173091888428,
|
| 408 |
+
"logps/chosen": -484.43389892578125,
|
| 409 |
+
"logps/rejected": -375.3249816894531,
|
| 410 |
+
"loss": 0.2936,
|
| 411 |
+
"rewards/accuracies": 0.706250011920929,
|
| 412 |
+
"rewards/chosen": -0.9159828424453735,
|
| 413 |
+
"rewards/margins": 0.4136818051338196,
|
| 414 |
+
"rewards/rejected": -1.3296645879745483,
|
| 415 |
"step": 280
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 0.81,
|
| 419 |
+
"learning_rate": 1.474239709917218e-06,
|
| 420 |
+
"logits/chosen": 0.05337507277727127,
|
| 421 |
+
"logits/rejected": 0.19436481595039368,
|
| 422 |
+
"logps/chosen": -443.375244140625,
|
| 423 |
+
"logps/rejected": -374.6485900878906,
|
| 424 |
+
"loss": 0.2803,
|
| 425 |
+
"rewards/accuracies": 0.612500011920929,
|
| 426 |
+
"rewards/chosen": -0.916093647480011,
|
| 427 |
+
"rewards/margins": 0.25273779034614563,
|
| 428 |
+
"rewards/rejected": -1.1688312292099,
|
| 429 |
"step": 290
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.84,
|
| 433 |
+
"learning_rate": 1.430400156338457e-06,
|
| 434 |
+
"logits/chosen": -0.04387836158275604,
|
| 435 |
+
"logits/rejected": 0.08865033835172653,
|
| 436 |
+
"logps/chosen": -432.93719482421875,
|
| 437 |
+
"logps/rejected": -336.19378662109375,
|
| 438 |
+
"loss": 0.2739,
|
| 439 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 440 |
+
"rewards/chosen": -0.5575471520423889,
|
| 441 |
+
"rewards/margins": 0.33055704832077026,
|
| 442 |
+
"rewards/rejected": -0.888104259967804,
|
| 443 |
"step": 300
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 0.87,
|
| 447 |
+
"learning_rate": 1.3855204839045892e-06,
|
| 448 |
+
"logits/chosen": -0.09471622109413147,
|
| 449 |
+
"logits/rejected": 0.04182841628789902,
|
| 450 |
+
"logps/chosen": -394.19952392578125,
|
| 451 |
+
"logps/rejected": -326.4076232910156,
|
| 452 |
+
"loss": 0.2828,
|
| 453 |
+
"rewards/accuracies": 0.65625,
|
| 454 |
+
"rewards/chosen": -0.5341824293136597,
|
| 455 |
+
"rewards/margins": 0.30772727727890015,
|
| 456 |
+
"rewards/rejected": -0.8419097661972046,
|
| 457 |
"step": 310
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.9,
|
| 461 |
+
"learning_rate": 1.3397091502748927e-06,
|
| 462 |
+
"logits/chosen": 0.06928735971450806,
|
| 463 |
+
"logits/rejected": 0.12053768336772919,
|
| 464 |
+
"logps/chosen": -407.7933654785156,
|
| 465 |
+
"logps/rejected": -357.62225341796875,
|
| 466 |
+
"loss": 0.3016,
|
| 467 |
+
"rewards/accuracies": 0.581250011920929,
|
| 468 |
+
"rewards/chosen": -0.7507076263427734,
|
| 469 |
+
"rewards/margins": 0.2525845170021057,
|
| 470 |
+
"rewards/rejected": -1.0032920837402344,
|
| 471 |
"step": 320
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 0.93,
|
| 475 |
+
"learning_rate": 1.2930768645910449e-06,
|
| 476 |
+
"logits/chosen": -0.04908103495836258,
|
| 477 |
+
"logits/rejected": 0.05275033786892891,
|
| 478 |
+
"logps/chosen": -489.91741943359375,
|
| 479 |
+
"logps/rejected": -391.38861083984375,
|
| 480 |
+
"loss": 0.269,
|
| 481 |
+
"rewards/accuracies": 0.625,
|
| 482 |
+
"rewards/chosen": -0.7464247941970825,
|
| 483 |
+
"rewards/margins": 0.3772323727607727,
|
| 484 |
+
"rewards/rejected": -1.1236572265625,
|
| 485 |
"step": 330
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"epoch": 0.96,
|
| 489 |
+
"learning_rate": 1.2457363199338495e-06,
|
| 490 |
+
"logits/chosen": -0.023084616288542747,
|
| 491 |
+
"logits/rejected": 0.06722521036863327,
|
| 492 |
+
"logps/chosen": -424.7259216308594,
|
| 493 |
+
"logps/rejected": -344.1465148925781,
|
| 494 |
+
"loss": 0.2641,
|
| 495 |
+
"rewards/accuracies": 0.643750011920929,
|
| 496 |
+
"rewards/chosen": -0.7394579648971558,
|
| 497 |
+
"rewards/margins": 0.2995792627334595,
|
| 498 |
+
"rewards/rejected": -1.0390373468399048,
|
| 499 |
"step": 340
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 0.98,
|
| 503 |
+
"learning_rate": 1.1978019209855173e-06,
|
| 504 |
+
"logits/chosen": -0.03397887200117111,
|
| 505 |
+
"logits/rejected": 0.0008539498085156083,
|
| 506 |
+
"logps/chosen": -443.153076171875,
|
| 507 |
+
"logps/rejected": -367.44195556640625,
|
| 508 |
+
"loss": 0.2896,
|
| 509 |
+
"rewards/accuracies": 0.668749988079071,
|
| 510 |
+
"rewards/chosen": -0.816065788269043,
|
| 511 |
+
"rewards/margins": 0.3647410273551941,
|
| 512 |
+
"rewards/rejected": -1.1808068752288818,
|
| 513 |
"step": 350
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 1.01,
|
| 517 |
+
"learning_rate": 1.14938950755563e-06,
|
| 518 |
+
"logits/chosen": -0.0028023570775985718,
|
| 519 |
+
"logits/rejected": 0.09356020390987396,
|
| 520 |
+
"logps/chosen": -409.0108947753906,
|
| 521 |
+
"logps/rejected": -419.5281677246094,
|
| 522 |
+
"loss": 0.2361,
|
| 523 |
+
"rewards/accuracies": 0.6937500238418579,
|
| 524 |
+
"rewards/chosen": -0.937981128692627,
|
| 525 |
+
"rewards/margins": 0.5299195051193237,
|
| 526 |
+
"rewards/rejected": -1.4679005146026611,
|
| 527 |
+
"step": 360
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.04,
|
| 531 |
+
"learning_rate": 1.1006160746389332e-06,
|
| 532 |
+
"logits/chosen": -0.15354487299919128,
|
| 533 |
+
"logits/rejected": -0.0770144909620285,
|
| 534 |
+
"logps/chosen": -461.51806640625,
|
| 535 |
+
"logps/rejected": -445.2923278808594,
|
| 536 |
+
"loss": 0.102,
|
| 537 |
+
"rewards/accuracies": 0.8125,
|
| 538 |
+
"rewards/chosen": -1.0436856746673584,
|
| 539 |
+
"rewards/margins": 0.894515335559845,
|
| 540 |
+
"rewards/rejected": -1.9382011890411377,
|
| 541 |
+
"step": 370
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.07,
|
| 545 |
+
"learning_rate": 1.0515994896814731e-06,
|
| 546 |
+
"logits/chosen": -0.12813356518745422,
|
| 547 |
+
"logits/rejected": -0.008146116510033607,
|
| 548 |
+
"logps/chosen": -474.5690002441406,
|
| 549 |
+
"logps/rejected": -505.58367919921875,
|
| 550 |
+
"loss": 0.076,
|
| 551 |
+
"rewards/accuracies": 0.800000011920929,
|
| 552 |
+
"rewards/chosen": -1.539747953414917,
|
| 553 |
+
"rewards/margins": 0.9511996507644653,
|
| 554 |
+
"rewards/rejected": -2.490947723388672,
|
| 555 |
+
"step": 380
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.1,
|
| 559 |
+
"learning_rate": 1.002458207738333e-06,
|
| 560 |
+
"logits/chosen": 0.014720706269145012,
|
| 561 |
+
"logits/rejected": 0.19409914314746857,
|
| 562 |
+
"logps/chosen": -462.5455627441406,
|
| 563 |
+
"logps/rejected": -508.26947021484375,
|
| 564 |
+
"loss": 0.0717,
|
| 565 |
+
"rewards/accuracies": 0.793749988079071,
|
| 566 |
+
"rewards/chosen": -1.7438398599624634,
|
| 567 |
+
"rewards/margins": 1.1367576122283936,
|
| 568 |
+
"rewards/rejected": -2.8805973529815674,
|
| 569 |
+
"step": 390
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.12,
|
| 573 |
+
"learning_rate": 9.533109852113413e-07,
|
| 574 |
+
"logits/chosen": 0.06460610777139664,
|
| 575 |
+
"logits/rejected": 0.15287192165851593,
|
| 576 |
+
"logps/chosen": -476.78753662109375,
|
| 577 |
+
"logps/rejected": -506.93701171875,
|
| 578 |
+
"loss": 0.0591,
|
| 579 |
+
"rewards/accuracies": 0.8062499761581421,
|
| 580 |
+
"rewards/chosen": -1.3379477262496948,
|
| 581 |
+
"rewards/margins": 1.1289650201797485,
|
| 582 |
+
"rewards/rejected": -2.4669127464294434,
|
| 583 |
+
"step": 400
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.15,
|
| 587 |
+
"learning_rate": 9.042765928585326e-07,
|
| 588 |
+
"logits/chosen": 0.21788999438285828,
|
| 589 |
+
"logits/rejected": 0.2529798150062561,
|
| 590 |
+
"logps/chosen": -483.2903747558594,
|
| 591 |
+
"logps/rejected": -527.7869262695312,
|
| 592 |
+
"loss": 0.0602,
|
| 593 |
+
"rewards/accuracies": 0.856249988079071,
|
| 594 |
+
"rewards/chosen": -1.615551233291626,
|
| 595 |
+
"rewards/margins": 1.4097917079925537,
|
| 596 |
+
"rewards/rejected": -3.0253429412841797,
|
| 597 |
+
"step": 410
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.18,
|
| 601 |
+
"learning_rate": 8.554735287689148e-07,
|
| 602 |
+
"logits/chosen": 0.1305566430091858,
|
| 603 |
+
"logits/rejected": 0.27758485078811646,
|
| 604 |
+
"logps/chosen": -527.9423217773438,
|
| 605 |
+
"logps/rejected": -558.6550903320312,
|
| 606 |
+
"loss": 0.062,
|
| 607 |
+
"rewards/accuracies": 0.768750011920929,
|
| 608 |
+
"rewards/chosen": -1.9209682941436768,
|
| 609 |
+
"rewards/margins": 1.1167800426483154,
|
| 610 |
+
"rewards/rejected": -3.037747859954834,
|
| 611 |
+
"step": 420
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.21,
|
| 615 |
+
"learning_rate": 8.070197319961782e-07,
|
| 616 |
+
"logits/chosen": 0.25931409001350403,
|
| 617 |
+
"logits/rejected": 0.3246735632419586,
|
| 618 |
+
"logps/chosen": -550.6995849609375,
|
| 619 |
+
"logps/rejected": -537.9232177734375,
|
| 620 |
+
"loss": 0.0486,
|
| 621 |
+
"rewards/accuracies": 0.8500000238418579,
|
| 622 |
+
"rewards/chosen": -1.6232093572616577,
|
| 623 |
+
"rewards/margins": 1.3964422941207886,
|
| 624 |
+
"rewards/rejected": -3.0196518898010254,
|
| 625 |
+
"step": 430
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.24,
|
| 629 |
+
"learning_rate": 7.590322975433856e-07,
|
| 630 |
+
"logits/chosen": 0.2402382791042328,
|
| 631 |
+
"logits/rejected": 0.37099042534828186,
|
| 632 |
+
"logps/chosen": -511.0347595214844,
|
| 633 |
+
"logps/rejected": -555.0653686523438,
|
| 634 |
+
"loss": 0.0533,
|
| 635 |
+
"rewards/accuracies": 0.78125,
|
| 636 |
+
"rewards/chosen": -1.9829241037368774,
|
| 637 |
+
"rewards/margins": 1.3394062519073486,
|
| 638 |
+
"rewards/rejected": -3.3223304748535156,
|
| 639 |
+
"step": 440
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.26,
|
| 643 |
+
"learning_rate": 7.116271933874245e-07,
|
| 644 |
+
"logits/chosen": 0.2019539177417755,
|
| 645 |
+
"logits/rejected": 0.3705137372016907,
|
| 646 |
+
"logps/chosen": -551.7188110351562,
|
| 647 |
+
"logps/rejected": -568.9718017578125,
|
| 648 |
+
"loss": 0.0512,
|
| 649 |
+
"rewards/accuracies": 0.8062499761581421,
|
| 650 |
+
"rewards/chosen": -1.8252284526824951,
|
| 651 |
+
"rewards/margins": 1.3718860149383545,
|
| 652 |
+
"rewards/rejected": -3.1971147060394287,
|
| 653 |
+
"step": 450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.29,
|
| 657 |
+
"learning_rate": 6.649189802270652e-07,
|
| 658 |
+
"logits/chosen": 0.2931487560272217,
|
| 659 |
+
"logits/rejected": 0.4406962990760803,
|
| 660 |
+
"logps/chosen": -505.7900390625,
|
| 661 |
+
"logps/rejected": -604.8612060546875,
|
| 662 |
+
"loss": 0.0489,
|
| 663 |
+
"rewards/accuracies": 0.793749988079071,
|
| 664 |
+
"rewards/chosen": -1.829275131225586,
|
| 665 |
+
"rewards/margins": 1.4683018922805786,
|
| 666 |
+
"rewards/rejected": -3.297576904296875,
|
| 667 |
+
"step": 460
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.32,
|
| 671 |
+
"learning_rate": 6.190205346318926e-07,
|
| 672 |
+
"logits/chosen": 0.15103109180927277,
|
| 673 |
+
"logits/rejected": 0.2090083360671997,
|
| 674 |
+
"logps/chosen": -548.1585693359375,
|
| 675 |
+
"logps/rejected": -613.2918701171875,
|
| 676 |
+
"loss": 0.0454,
|
| 677 |
+
"rewards/accuracies": 0.8187500238418579,
|
| 678 |
+
"rewards/chosen": -2.007948398590088,
|
| 679 |
+
"rewards/margins": 1.6015655994415283,
|
| 680 |
+
"rewards/rejected": -3.609513759613037,
|
| 681 |
+
"step": 470
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.35,
|
| 685 |
+
"learning_rate": 5.740427762611604e-07,
|
| 686 |
+
"logits/chosen": 0.24043090641498566,
|
| 687 |
+
"logits/rejected": 0.37472882866859436,
|
| 688 |
+
"logps/chosen": -557.3897705078125,
|
| 689 |
+
"logps/rejected": -610.3823852539062,
|
| 690 |
+
"loss": 0.0382,
|
| 691 |
+
"rewards/accuracies": 0.8125,
|
| 692 |
+
"rewards/chosen": -2.103377103805542,
|
| 693 |
+
"rewards/margins": 1.6295030117034912,
|
| 694 |
+
"rewards/rejected": -3.732880115509033,
|
| 695 |
+
"step": 480
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.38,
|
| 699 |
+
"learning_rate": 5.300943998117749e-07,
|
| 700 |
+
"logits/chosen": 0.23212356865406036,
|
| 701 |
+
"logits/rejected": 0.3468802273273468,
|
| 702 |
+
"logps/chosen": -589.6870727539062,
|
| 703 |
+
"logps/rejected": -632.7547607421875,
|
| 704 |
+
"loss": 0.0425,
|
| 705 |
+
"rewards/accuracies": 0.84375,
|
| 706 |
+
"rewards/chosen": -2.1241652965545654,
|
| 707 |
+
"rewards/margins": 1.638109564781189,
|
| 708 |
+
"rewards/rejected": -3.762274980545044,
|
| 709 |
+
"step": 490
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.4,
|
| 713 |
+
"learning_rate": 4.872816123431976e-07,
|
| 714 |
+
"logits/chosen": 0.17420414090156555,
|
| 715 |
+
"logits/rejected": 0.3029765188694,
|
| 716 |
+
"logps/chosen": -583.6693115234375,
|
| 717 |
+
"logps/rejected": -632.7525024414062,
|
| 718 |
+
"loss": 0.0332,
|
| 719 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 720 |
+
"rewards/chosen": -2.0016562938690186,
|
| 721 |
+
"rewards/margins": 1.703657865524292,
|
| 722 |
+
"rewards/rejected": -3.7053139209747314,
|
| 723 |
+
"step": 500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.43,
|
| 727 |
+
"learning_rate": 4.4570787661405e-07,
|
| 728 |
+
"logits/chosen": 0.24483375251293182,
|
| 729 |
+
"logits/rejected": 0.43443599343299866,
|
| 730 |
+
"logps/chosen": -568.5266723632812,
|
| 731 |
+
"logps/rejected": -630.6636962890625,
|
| 732 |
+
"loss": 0.0398,
|
| 733 |
+
"rewards/accuracies": 0.800000011920929,
|
| 734 |
+
"rewards/chosen": -2.0844273567199707,
|
| 735 |
+
"rewards/margins": 1.567047357559204,
|
| 736 |
+
"rewards/rejected": -3.651474714279175,
|
| 737 |
+
"step": 510
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.46,
|
| 741 |
+
"learning_rate": 4.0547366105068347e-07,
|
| 742 |
+
"logits/chosen": 0.2337774932384491,
|
| 743 |
+
"logits/rejected": 0.3223033547401428,
|
| 744 |
+
"logps/chosen": -556.1533813476562,
|
| 745 |
+
"logps/rejected": -626.2739868164062,
|
| 746 |
+
"loss": 0.041,
|
| 747 |
+
"rewards/accuracies": 0.831250011920929,
|
| 748 |
+
"rewards/chosen": -2.1958367824554443,
|
| 749 |
+
"rewards/margins": 1.7631685733795166,
|
| 750 |
+
"rewards/rejected": -3.959005355834961,
|
| 751 |
+
"step": 520
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.49,
|
| 755 |
+
"learning_rate": 3.666761969519528e-07,
|
| 756 |
+
"logits/chosen": 0.2308119833469391,
|
| 757 |
+
"logits/rejected": 0.40482252836227417,
|
| 758 |
+
"logps/chosen": -531.3250732421875,
|
| 759 |
+
"logps/rejected": -586.0523071289062,
|
| 760 |
+
"loss": 0.0321,
|
| 761 |
+
"rewards/accuracies": 0.78125,
|
| 762 |
+
"rewards/chosen": -1.9775184392929077,
|
| 763 |
+
"rewards/margins": 1.4050841331481934,
|
| 764 |
+
"rewards/rejected": -3.3826022148132324,
|
| 765 |
+
"step": 530
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.52,
|
| 769 |
+
"learning_rate": 3.2940924351693213e-07,
|
| 770 |
+
"logits/chosen": 0.18636593222618103,
|
| 771 |
+
"logits/rejected": 0.3547353744506836,
|
| 772 |
+
"logps/chosen": -605.5850830078125,
|
| 773 |
+
"logps/rejected": -643.3690185546875,
|
| 774 |
+
"loss": 0.0335,
|
| 775 |
+
"rewards/accuracies": 0.8500000238418579,
|
| 776 |
+
"rewards/chosen": -2.183945894241333,
|
| 777 |
+
"rewards/margins": 1.709857702255249,
|
| 778 |
+
"rewards/rejected": -3.8938040733337402,
|
| 779 |
+
"step": 540
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.55,
|
| 783 |
+
"learning_rate": 2.937628612634184e-07,
|
| 784 |
+
"logits/chosen": 0.2420307844877243,
|
| 785 |
+
"logits/rejected": 0.41368550062179565,
|
| 786 |
+
"logps/chosen": -569.5245971679688,
|
| 787 |
+
"logps/rejected": -632.2020263671875,
|
| 788 |
+
"loss": 0.0344,
|
| 789 |
+
"rewards/accuracies": 0.8187500238418579,
|
| 790 |
+
"rewards/chosen": -2.410912036895752,
|
| 791 |
+
"rewards/margins": 1.8822616338729858,
|
| 792 |
+
"rewards/rejected": -4.293173789978027,
|
| 793 |
+
"step": 550
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.57,
|
| 797 |
+
"learning_rate": 2.598231943847916e-07,
|
| 798 |
+
"logits/chosen": 0.22710604965686798,
|
| 799 |
+
"logits/rejected": 0.4254523813724518,
|
| 800 |
+
"logps/chosen": -606.672119140625,
|
| 801 |
+
"logps/rejected": -696.2296142578125,
|
| 802 |
+
"loss": 0.0376,
|
| 803 |
+
"rewards/accuracies": 0.78125,
|
| 804 |
+
"rewards/chosen": -2.5400805473327637,
|
| 805 |
+
"rewards/margins": 1.9197076559066772,
|
| 806 |
+
"rewards/rejected": -4.4597883224487305,
|
| 807 |
+
"step": 560
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.6,
|
| 811 |
+
"learning_rate": 2.276722625711861e-07,
|
| 812 |
+
"logits/chosen": 0.2909803092479706,
|
| 813 |
+
"logits/rejected": 0.4825025200843811,
|
| 814 |
+
"logps/chosen": -583.4527587890625,
|
| 815 |
+
"logps/rejected": -639.517333984375,
|
| 816 |
+
"loss": 0.0302,
|
| 817 |
+
"rewards/accuracies": 0.8500000238418579,
|
| 818 |
+
"rewards/chosen": -2.0665905475616455,
|
| 819 |
+
"rewards/margins": 1.7480655908584595,
|
| 820 |
+
"rewards/rejected": -3.8146564960479736,
|
| 821 |
+
"step": 570
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.63,
|
| 825 |
+
"learning_rate": 1.973877627980699e-07,
|
| 826 |
+
"logits/chosen": 0.26504355669021606,
|
| 827 |
+
"logits/rejected": 0.332762211561203,
|
| 828 |
+
"logps/chosen": -514.5237426757812,
|
| 829 |
+
"logps/rejected": -629.0364990234375,
|
| 830 |
+
"loss": 0.0301,
|
| 831 |
+
"rewards/accuracies": 0.8187500238418579,
|
| 832 |
+
"rewards/chosen": -1.9947683811187744,
|
| 833 |
+
"rewards/margins": 1.5820239782333374,
|
| 834 |
+
"rewards/rejected": -3.5767929553985596,
|
| 835 |
+
"step": 580
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.66,
|
| 839 |
+
"learning_rate": 1.6904288156123636e-07,
|
| 840 |
+
"logits/chosen": 0.23839843273162842,
|
| 841 |
+
"logits/rejected": 0.39869600534439087,
|
| 842 |
+
"logps/chosen": -580.44921875,
|
| 843 |
+
"logps/rejected": -675.9835815429688,
|
| 844 |
+
"loss": 0.0358,
|
| 845 |
+
"rewards/accuracies": 0.856249988079071,
|
| 846 |
+
"rewards/chosen": -2.2089920043945312,
|
| 847 |
+
"rewards/margins": 1.9142862558364868,
|
| 848 |
+
"rewards/rejected": -4.12327766418457,
|
| 849 |
+
"step": 590
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.69,
|
| 853 |
+
"learning_rate": 1.4270611801196642e-07,
|
| 854 |
+
"logits/chosen": 0.32313600182533264,
|
| 855 |
+
"logits/rejected": 0.4433811604976654,
|
| 856 |
+
"logps/chosen": -588.3478393554688,
|
| 857 |
+
"logps/rejected": -604.0819091796875,
|
| 858 |
+
"loss": 0.031,
|
| 859 |
+
"rewards/accuracies": 0.793749988079071,
|
| 860 |
+
"rewards/chosen": -2.1629788875579834,
|
| 861 |
+
"rewards/margins": 1.6147849559783936,
|
| 862 |
+
"rewards/rejected": -3.777763843536377,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.71,
|
| 867 |
+
"learning_rate": 1.1844111841977633e-07,
|
| 868 |
+
"logits/chosen": 0.2681792378425598,
|
| 869 |
+
"logits/rejected": 0.34347304701805115,
|
| 870 |
+
"logps/chosen": -597.367431640625,
|
| 871 |
+
"logps/rejected": -694.4761962890625,
|
| 872 |
+
"loss": 0.0319,
|
| 873 |
+
"rewards/accuracies": 0.8187500238418579,
|
| 874 |
+
"rewards/chosen": -2.333829879760742,
|
| 875 |
+
"rewards/margins": 1.9641401767730713,
|
| 876 |
+
"rewards/rejected": -4.297970294952393,
|
| 877 |
+
"step": 610
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.74,
|
| 881 |
+
"learning_rate": 9.630652236279625e-08,
|
| 882 |
+
"logits/chosen": 0.28177735209465027,
|
| 883 |
+
"logits/rejected": 0.3899393081665039,
|
| 884 |
+
"logps/chosen": -565.4825439453125,
|
| 885 |
+
"logps/rejected": -616.7823486328125,
|
| 886 |
+
"loss": 0.0329,
|
| 887 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 888 |
+
"rewards/chosen": -2.153439521789551,
|
| 889 |
+
"rewards/margins": 1.7681152820587158,
|
| 890 |
+
"rewards/rejected": -3.9215550422668457,
|
| 891 |
+
"step": 620
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.77,
|
| 895 |
+
"learning_rate": 7.63558210174814e-08,
|
| 896 |
+
"logits/chosen": 0.21648459136486053,
|
| 897 |
+
"logits/rejected": 0.4712795615196228,
|
| 898 |
+
"logps/chosen": -561.133056640625,
|
| 899 |
+
"logps/rejected": -631.644287109375,
|
| 900 |
+
"loss": 0.0334,
|
| 901 |
+
"rewards/accuracies": 0.800000011920929,
|
| 902 |
+
"rewards/chosen": -2.387845754623413,
|
| 903 |
+
"rewards/margins": 1.6112064123153687,
|
| 904 |
+
"rewards/rejected": -3.9990525245666504,
|
| 905 |
+
"step": 630
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.8,
|
| 909 |
+
"learning_rate": 5.8637227890115273e-08,
|
| 910 |
+
"logits/chosen": 0.23778888583183289,
|
| 911 |
+
"logits/rejected": 0.34435296058654785,
|
| 912 |
+
"logps/chosen": -649.9002685546875,
|
| 913 |
+
"logps/rejected": -715.91943359375,
|
| 914 |
+
"loss": 0.0299,
|
| 915 |
+
"rewards/accuracies": 0.856249988079071,
|
| 916 |
+
"rewards/chosen": -2.4873223304748535,
|
| 917 |
+
"rewards/margins": 2.0735325813293457,
|
| 918 |
+
"rewards/rejected": -4.560854911804199,
|
| 919 |
+
"step": 640
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.83,
|
| 923 |
+
"learning_rate": 4.3193562302499046e-08,
|
| 924 |
+
"logits/chosen": 0.3359209895133972,
|
| 925 |
+
"logits/rejected": 0.4597659707069397,
|
| 926 |
+
"logps/chosen": -579.0816650390625,
|
| 927 |
+
"logps/rejected": -621.2535400390625,
|
| 928 |
+
"loss": 0.0336,
|
| 929 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 930 |
+
"rewards/chosen": -2.270595073699951,
|
| 931 |
+
"rewards/margins": 1.7250922918319702,
|
| 932 |
+
"rewards/rejected": -3.995687484741211,
|
| 933 |
+
"step": 650
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.85,
|
| 937 |
+
"learning_rate": 3.006214591340339e-08,
|
| 938 |
+
"logits/chosen": 0.2430475950241089,
|
| 939 |
+
"logits/rejected": 0.3344910740852356,
|
| 940 |
+
"logps/chosen": -618.7166748046875,
|
| 941 |
+
"logps/rejected": -714.7845458984375,
|
| 942 |
+
"loss": 0.0302,
|
| 943 |
+
"rewards/accuracies": 0.8687499761581421,
|
| 944 |
+
"rewards/chosen": -2.330672264099121,
|
| 945 |
+
"rewards/margins": 2.0827202796936035,
|
| 946 |
+
"rewards/rejected": -4.413392543792725,
|
| 947 |
+
"step": 660
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.88,
|
| 951 |
+
"learning_rate": 1.9274712525847447e-08,
|
| 952 |
+
"logits/chosen": 0.3407454490661621,
|
| 953 |
+
"logits/rejected": 0.4727950692176819,
|
| 954 |
+
"logps/chosen": -568.6539306640625,
|
| 955 |
+
"logps/rejected": -646.0457153320312,
|
| 956 |
+
"loss": 0.0271,
|
| 957 |
+
"rewards/accuracies": 0.8125,
|
| 958 |
+
"rewards/chosen": -2.3302292823791504,
|
| 959 |
+
"rewards/margins": 1.696593999862671,
|
| 960 |
+
"rewards/rejected": -4.0268235206604,
|
| 961 |
+
"step": 670
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.91,
|
| 965 |
+
"learning_rate": 1.0857331398169577e-08,
|
| 966 |
+
"logits/chosen": 0.16540665924549103,
|
| 967 |
+
"logits/rejected": 0.4224317669868469,
|
| 968 |
+
"logps/chosen": -581.6246337890625,
|
| 969 |
+
"logps/rejected": -642.5072021484375,
|
| 970 |
+
"loss": 0.0299,
|
| 971 |
+
"rewards/accuracies": 0.824999988079071,
|
| 972 |
+
"rewards/chosen": -2.267688274383545,
|
| 973 |
+
"rewards/margins": 1.918178915977478,
|
| 974 |
+
"rewards/rejected": -4.185866832733154,
|
| 975 |
+
"step": 680
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.94,
|
| 979 |
+
"learning_rate": 4.830344244220686e-09,
|
| 980 |
+
"logits/chosen": 0.14440816640853882,
|
| 981 |
+
"logits/rejected": 0.27977341413497925,
|
| 982 |
+
"logps/chosen": -637.5778198242188,
|
| 983 |
+
"logps/rejected": -684.380859375,
|
| 984 |
+
"loss": 0.0277,
|
| 985 |
+
"rewards/accuracies": 0.8687499761581421,
|
| 986 |
+
"rewards/chosen": -2.4349887371063232,
|
| 987 |
+
"rewards/margins": 1.7913310527801514,
|
| 988 |
+
"rewards/rejected": -4.226320266723633,
|
| 989 |
+
"step": 690
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.97,
|
| 993 |
+
"learning_rate": 1.2083160749236653e-09,
|
| 994 |
+
"logits/chosen": 0.23438580334186554,
|
| 995 |
+
"logits/rejected": 0.3880893290042877,
|
| 996 |
+
"logps/chosen": -624.1004028320312,
|
| 997 |
+
"logps/rejected": -729.334716796875,
|
| 998 |
+
"loss": 0.0276,
|
| 999 |
+
"rewards/accuracies": 0.831250011920929,
|
| 1000 |
+
"rewards/chosen": -2.6064438819885254,
|
| 1001 |
+
"rewards/margins": 1.8713786602020264,
|
| 1002 |
+
"rewards/rejected": -4.477822780609131,
|
| 1003 |
+
"step": 700
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.0,
|
| 1007 |
+
"learning_rate": 0.0,
|
| 1008 |
+
"logits/chosen": 0.3188737630844116,
|
| 1009 |
+
"logits/rejected": 0.4447614252567291,
|
| 1010 |
+
"logps/chosen": -564.446533203125,
|
| 1011 |
+
"logps/rejected": -686.6253662109375,
|
| 1012 |
+
"loss": 0.0308,
|
| 1013 |
+
"rewards/accuracies": 0.887499988079071,
|
| 1014 |
+
"rewards/chosen": -2.405637264251709,
|
| 1015 |
+
"rewards/margins": 1.9922983646392822,
|
| 1016 |
+
"rewards/rejected": -4.39793586730957,
|
| 1017 |
+
"step": 710
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.0,
|
| 1021 |
+
"step": 710,
|
| 1022 |
"total_flos": 0.0,
|
| 1023 |
+
"train_loss": 0.18113358499298632,
|
| 1024 |
+
"train_runtime": 8124.3794,
|
| 1025 |
+
"train_samples_per_second": 11.213,
|
| 1026 |
"train_steps_per_second": 0.087
|
| 1027 |
}
|
| 1028 |
],
|
| 1029 |
"logging_steps": 10,
|
| 1030 |
+
"max_steps": 710,
|
| 1031 |
+
"num_train_epochs": 2,
|
| 1032 |
"save_steps": 10000,
|
| 1033 |
"total_flos": 0.0,
|
| 1034 |
"trial_name": null,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6008
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fda369e142d3189323c2d59bf379d7fe0026912ef983c553220d80f8569efb58
|
| 3 |
size 6008
|