Model save
Browse files- README.md +2 -2
- 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 +361 -865
- 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: 1e-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: 1
|
| 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":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.4252443082735572,
|
| 4 |
+
"train_runtime": 5295.2592,
|
| 5 |
"train_samples": 45548,
|
| 6 |
+
"train_samples_per_second": 8.602,
|
| 7 |
+
"train_steps_per_second": 0.067
|
| 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:dd6c8b5e789d40ac409360c424a8b67feda543a63781afe42e29d899d8d9eb06
|
| 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 sha256:5b0a1631656d2ecd37e77ff857273ec92c8d505597cd8bfa395387a354a8e212
|
| 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:f3e1540e1e486049181a6f16931c2afcd0a03d7b78d6e2761ccc32a222ceeea9
|
| 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 sha256:05e6a30991b3a5a7c162f3ac654d34e3fae77ecb769c5715beb974b7faaf2c0f
|
| 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":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.4252443082735572,
|
| 4 |
+
"train_runtime": 5295.2592,
|
| 5 |
"train_samples": 45548,
|
| 6 |
+
"train_samples_per_second": 8.602,
|
| 7 |
+
"train_steps_per_second": 0.067
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,21 +1,21 @@
|
|
| 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.
|
| 15 |
-
"logits/rejected": -0.
|
| 16 |
-
"logps/chosen": -
|
| 17 |
-
"logps/rejected": -
|
| 18 |
-
"loss": 0.
|
| 19 |
"rewards/accuracies": 0.0,
|
| 20 |
"rewards/chosen": 0.0,
|
| 21 |
"rewards/margins": 0.0,
|
|
@@ -24,1011 +24,507 @@
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03,
|
| 27 |
-
"learning_rate":
|
| 28 |
-
"logits/chosen": -0.
|
| 29 |
-
"logits/rejected": -0.
|
| 30 |
-
"logps/chosen": -
|
| 31 |
-
"logps/rejected": -
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"rewards/accuracies": 0.
|
| 34 |
-
"rewards/chosen": -0.
|
| 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": -
|
| 45 |
-
"logps/rejected": -
|
| 46 |
-
"loss": 0.
|
| 47 |
-
"rewards/accuracies": 0.
|
| 48 |
-
"rewards/chosen":
|
| 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": -
|
| 59 |
-
"logps/rejected": -
|
| 60 |
-
"loss": 0.
|
| 61 |
-
"rewards/accuracies": 0.
|
| 62 |
-
"rewards/chosen":
|
| 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": 0.5687500238418579,
|
| 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": 0.71875,
|
| 146 |
-
"rewards/chosen":
|
| 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":
|
| 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":
|
| 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":
|
| 189 |
-
"rewards/margins": 0.
|
| 190 |
-
"rewards/rejected": -
|
| 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":
|
| 203 |
-
"rewards/margins": 0.
|
| 204 |
-
"rewards/rejected": -
|
| 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":
|
| 217 |
-
"rewards/margins": 0.
|
| 218 |
-
"rewards/rejected": -
|
| 219 |
"step": 140
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.42,
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"logits/chosen": 0.
|
| 225 |
-
"logits/rejected": 0.
|
| 226 |
-
"logps/chosen": -
|
| 227 |
-
"logps/rejected": -
|
| 228 |
-
"loss": 0.
|
| 229 |
-
"rewards/accuracies": 0.
|
| 230 |
-
"rewards/chosen": -
|
| 231 |
-
"rewards/margins": 0.
|
| 232 |
-
"rewards/rejected": -
|
| 233 |
"step": 150
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.45,
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"logits/chosen": 0.
|
| 239 |
-
"logits/rejected": 0.
|
| 240 |
-
"logps/chosen": -
|
| 241 |
-
"logps/rejected": -
|
| 242 |
-
"loss": 0.
|
| 243 |
-
"rewards/accuracies": 0.
|
| 244 |
-
"rewards/chosen":
|
| 245 |
-
"rewards/margins": 0.
|
| 246 |
-
"rewards/rejected": -
|
| 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": -
|
| 259 |
-
"rewards/margins": 0.
|
| 260 |
-
"rewards/rejected": -
|
| 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": -
|
| 273 |
-
"rewards/margins": 0.
|
| 274 |
-
"rewards/rejected": -
|
| 275 |
"step": 180
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.53,
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"logits/chosen": 0.
|
| 281 |
-
"logits/rejected": 0.
|
| 282 |
-
"logps/chosen": -
|
| 283 |
-
"logps/rejected": -
|
| 284 |
-
"loss": 0.
|
| 285 |
"rewards/accuracies": 0.668749988079071,
|
| 286 |
-
"rewards/chosen":
|
| 287 |
-
"rewards/margins": 0.
|
| 288 |
-
"rewards/rejected": -
|
| 289 |
"step": 190
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.56,
|
| 293 |
-
"learning_rate": 4.
|
| 294 |
-
"logits/chosen": -0.
|
| 295 |
-
"logits/rejected": 0.
|
| 296 |
-
"logps/chosen": -
|
| 297 |
-
"logps/rejected": -
|
| 298 |
-
"loss": 0.
|
| 299 |
-
"rewards/accuracies": 0.
|
| 300 |
-
"rewards/chosen": -
|
| 301 |
-
"rewards/margins": 0.
|
| 302 |
-
"rewards/rejected": -
|
| 303 |
"step": 200
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.59,
|
| 307 |
-
"learning_rate": 4.
|
| 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": 0.
|
| 316 |
-
"rewards/rejected": -
|
| 317 |
"step": 210
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.62,
|
| 321 |
-
"learning_rate":
|
| 322 |
-
"logits/chosen": -0.
|
| 323 |
-
"logits/rejected": 0.
|
| 324 |
-
"logps/chosen": -
|
| 325 |
-
"logps/rejected": -
|
| 326 |
-
"loss": 0.
|
| 327 |
-
"rewards/accuracies": 0.
|
| 328 |
-
"rewards/chosen": -
|
| 329 |
-
"rewards/margins": 0.
|
| 330 |
-
"rewards/rejected": -
|
| 331 |
"step": 220
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.65,
|
| 335 |
-
"learning_rate":
|
| 336 |
-
"logits/chosen": 0.
|
| 337 |
-
"logits/rejected": 0.
|
| 338 |
-
"logps/chosen": -
|
| 339 |
-
"logps/rejected": -
|
| 340 |
-
"loss": 0.
|
| 341 |
-
"rewards/accuracies": 0.
|
| 342 |
-
"rewards/chosen": -
|
| 343 |
-
"rewards/margins": 0.
|
| 344 |
-
"rewards/rejected": -
|
| 345 |
"step": 230
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.67,
|
| 349 |
-
"learning_rate":
|
| 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": 0.
|
| 358 |
-
"rewards/rejected": -
|
| 359 |
"step": 240
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.7,
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"logits/chosen":
|
| 365 |
-
"logits/rejected": 0.
|
| 366 |
-
"logps/chosen": -
|
| 367 |
-
"logps/rejected": -
|
| 368 |
-
"loss": 0.
|
| 369 |
"rewards/accuracies": 0.6625000238418579,
|
| 370 |
-
"rewards/chosen": -
|
| 371 |
-
"rewards/margins": 0.
|
| 372 |
-
"rewards/rejected": -
|
| 373 |
"step": 250
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.73,
|
| 377 |
-
"learning_rate":
|
| 378 |
-
"logits/chosen": 0.
|
| 379 |
-
"logits/rejected": 0.
|
| 380 |
-
"logps/chosen": -
|
| 381 |
-
"logps/rejected": -
|
| 382 |
-
"loss": 0.
|
| 383 |
"rewards/accuracies": 0.625,
|
| 384 |
-
"rewards/chosen": -
|
| 385 |
-
"rewards/margins": 0.
|
| 386 |
-
"rewards/rejected": -
|
| 387 |
"step": 260
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.76,
|
| 391 |
-
"learning_rate":
|
| 392 |
-
"logits/chosen": 0.
|
| 393 |
-
"logits/rejected": 0.
|
| 394 |
-
"logps/chosen": -
|
| 395 |
-
"logps/rejected": -
|
| 396 |
-
"loss": 0.
|
| 397 |
-
"rewards/accuracies": 0.
|
| 398 |
-
"rewards/chosen": -
|
| 399 |
-
"rewards/margins": 0.
|
| 400 |
-
"rewards/rejected": -
|
| 401 |
"step": 270
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.79,
|
| 405 |
-
"learning_rate":
|
| 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": 0.
|
| 414 |
-
"rewards/rejected": -
|
| 415 |
"step": 280
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 0.81,
|
| 419 |
-
"learning_rate":
|
| 420 |
-
"logits/chosen": 0.
|
| 421 |
-
"logits/rejected": 0.
|
| 422 |
-
"logps/chosen": -
|
| 423 |
-
"logps/rejected": -
|
| 424 |
-
"loss": 0.
|
| 425 |
-
"rewards/accuracies": 0.
|
| 426 |
-
"rewards/chosen": -
|
| 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": -
|
| 441 |
-
"rewards/margins": 0.
|
| 442 |
-
"rewards/rejected": -
|
| 443 |
"step": 300
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 0.87,
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"logits/chosen": 0.
|
| 449 |
-
"logits/rejected": 0.
|
| 450 |
-
"logps/chosen": -
|
| 451 |
-
"logps/rejected": -
|
| 452 |
-
"loss": 0.
|
| 453 |
-
"rewards/accuracies": 0.
|
| 454 |
-
"rewards/chosen": -
|
| 455 |
-
"rewards/margins": 0.
|
| 456 |
-
"rewards/rejected": -
|
| 457 |
"step": 310
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.9,
|
| 461 |
-
"learning_rate":
|
| 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": 0.
|
| 470 |
-
"rewards/rejected": -
|
| 471 |
"step": 320
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 0.93,
|
| 475 |
-
"learning_rate":
|
| 476 |
-
"logits/chosen": 0.
|
| 477 |
-
"logits/rejected": 0.
|
| 478 |
-
"logps/chosen": -
|
| 479 |
-
"logps/rejected": -
|
| 480 |
-
"loss": 0.
|
| 481 |
-
"rewards/accuracies": 0.
|
| 482 |
-
"rewards/chosen": -
|
| 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": 0.
|
| 492 |
-
"logps/chosen": -
|
| 493 |
-
"logps/rejected": -
|
| 494 |
-
"loss": 0.
|
| 495 |
-
"rewards/accuracies": 0.
|
| 496 |
-
"rewards/chosen": -
|
| 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": 0.
|
| 506 |
-
"logps/chosen": -
|
| 507 |
-
"logps/rejected": -
|
| 508 |
-
"loss": 0.
|
| 509 |
-
"rewards/accuracies": 0.6812499761581421,
|
| 510 |
-
"rewards/chosen": -1.722198247909546,
|
| 511 |
-
"rewards/margins": 0.3744170069694519,
|
| 512 |
-
"rewards/rejected": -2.0966150760650635,
|
| 513 |
-
"step": 350
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 1.01,
|
| 517 |
-
"learning_rate": 2.8734737688890752e-06,
|
| 518 |
-
"logits/chosen": 0.3470906615257263,
|
| 519 |
-
"logits/rejected": 0.42189717292785645,
|
| 520 |
-
"logps/chosen": -492.981689453125,
|
| 521 |
-
"logps/rejected": -519.1780395507812,
|
| 522 |
-
"loss": 0.1261,
|
| 523 |
"rewards/accuracies": 0.6625000238418579,
|
| 524 |
-
"rewards/chosen": -
|
| 525 |
-
"rewards/margins": 0.
|
| 526 |
-
"rewards/rejected": -
|
| 527 |
-
"step":
|
| 528 |
-
},
|
| 529 |
-
{
|
| 530 |
-
"epoch": 1.04,
|
| 531 |
-
"learning_rate": 2.7515401865973333e-06,
|
| 532 |
-
"logits/chosen": 0.29653865098953247,
|
| 533 |
-
"logits/rejected": 0.31189900636672974,
|
| 534 |
-
"logps/chosen": -541.0857543945312,
|
| 535 |
-
"logps/rejected": -558.9527587890625,
|
| 536 |
-
"loss": 0.0325,
|
| 537 |
-
"rewards/accuracies": 0.8187500238418579,
|
| 538 |
-
"rewards/chosen": -1.839362382888794,
|
| 539 |
-
"rewards/margins": 1.2354429960250854,
|
| 540 |
-
"rewards/rejected": -3.074805498123169,
|
| 541 |
-
"step": 370
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 1.07,
|
| 545 |
-
"learning_rate": 2.628998724203683e-06,
|
| 546 |
-
"logits/chosen": 0.24393782019615173,
|
| 547 |
-
"logits/rejected": 0.3541429042816162,
|
| 548 |
-
"logps/chosen": -601.4793090820312,
|
| 549 |
-
"logps/rejected": -695.1011352539062,
|
| 550 |
-
"loss": 0.0162,
|
| 551 |
-
"rewards/accuracies": 0.793749988079071,
|
| 552 |
-
"rewards/chosen": -2.808850049972534,
|
| 553 |
-
"rewards/margins": 1.5772719383239746,
|
| 554 |
-
"rewards/rejected": -4.386122226715088,
|
| 555 |
-
"step": 380
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 1.1,
|
| 559 |
-
"learning_rate": 2.5061455193458326e-06,
|
| 560 |
-
"logits/chosen": 0.2903648018836975,
|
| 561 |
-
"logits/rejected": 0.4108067452907562,
|
| 562 |
-
"logps/chosen": -613.734619140625,
|
| 563 |
-
"logps/rejected": -712.7540893554688,
|
| 564 |
-
"loss": 0.0109,
|
| 565 |
-
"rewards/accuracies": 0.7749999761581421,
|
| 566 |
-
"rewards/chosen": -3.2557308673858643,
|
| 567 |
-
"rewards/margins": 1.669713020324707,
|
| 568 |
-
"rewards/rejected": -4.92544412612915,
|
| 569 |
-
"step": 390
|
| 570 |
-
},
|
| 571 |
-
{
|
| 572 |
-
"epoch": 1.12,
|
| 573 |
-
"learning_rate": 2.3832774630283534e-06,
|
| 574 |
-
"logits/chosen": 0.3819536864757538,
|
| 575 |
-
"logits/rejected": 0.4252726137638092,
|
| 576 |
-
"logps/chosen": -615.9334716796875,
|
| 577 |
-
"logps/rejected": -717.3001098632812,
|
| 578 |
-
"loss": 0.0112,
|
| 579 |
-
"rewards/accuracies": 0.8125,
|
| 580 |
-
"rewards/chosen": -2.7294070720672607,
|
| 581 |
-
"rewards/margins": 1.8411362171173096,
|
| 582 |
-
"rewards/rejected": -4.57054328918457,
|
| 583 |
-
"step": 400
|
| 584 |
-
},
|
| 585 |
-
{
|
| 586 |
-
"epoch": 1.15,
|
| 587 |
-
"learning_rate": 2.2606914821463317e-06,
|
| 588 |
-
"logits/chosen": 0.5641288161277771,
|
| 589 |
-
"logits/rejected": 0.5737628936767578,
|
| 590 |
-
"logps/chosen": -588.9856567382812,
|
| 591 |
-
"logps/rejected": -711.607421875,
|
| 592 |
-
"loss": 0.0108,
|
| 593 |
-
"rewards/accuracies": 0.793749988079071,
|
| 594 |
-
"rewards/chosen": -2.6725032329559326,
|
| 595 |
-
"rewards/margins": 2.191044330596924,
|
| 596 |
-
"rewards/rejected": -4.8635478019714355,
|
| 597 |
-
"step": 410
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 1.18,
|
| 601 |
-
"learning_rate": 2.138683821922287e-06,
|
| 602 |
-
"logits/chosen": 0.44424986839294434,
|
| 603 |
-
"logits/rejected": 0.5702834129333496,
|
| 604 |
-
"logps/chosen": -649.6632080078125,
|
| 605 |
-
"logps/rejected": -763.0523681640625,
|
| 606 |
-
"loss": 0.0084,
|
| 607 |
-
"rewards/accuracies": 0.8187500238418579,
|
| 608 |
-
"rewards/chosen": -3.1381773948669434,
|
| 609 |
-
"rewards/margins": 1.943543791770935,
|
| 610 |
-
"rewards/rejected": -5.081721305847168,
|
| 611 |
-
"step": 420
|
| 612 |
-
},
|
| 613 |
-
{
|
| 614 |
-
"epoch": 1.21,
|
| 615 |
-
"learning_rate": 2.0175493299904456e-06,
|
| 616 |
-
"logits/chosen": 0.5278526544570923,
|
| 617 |
-
"logits/rejected": 0.5699671506881714,
|
| 618 |
-
"logps/chosen": -760.1556396484375,
|
| 619 |
-
"logps/rejected": -837.56298828125,
|
| 620 |
-
"loss": 0.0057,
|
| 621 |
-
"rewards/accuracies": 0.8500000238418579,
|
| 622 |
-
"rewards/chosen": -3.7177700996398926,
|
| 623 |
-
"rewards/margins": 2.2982795238494873,
|
| 624 |
-
"rewards/rejected": -6.016049385070801,
|
| 625 |
-
"step": 430
|
| 626 |
-
},
|
| 627 |
-
{
|
| 628 |
-
"epoch": 1.24,
|
| 629 |
-
"learning_rate": 1.8975807438584643e-06,
|
| 630 |
-
"logits/chosen": 0.4808652400970459,
|
| 631 |
-
"logits/rejected": 0.5634757280349731,
|
| 632 |
-
"logps/chosen": -693.7796020507812,
|
| 633 |
-
"logps/rejected": -791.9578247070312,
|
| 634 |
-
"loss": 0.0054,
|
| 635 |
-
"rewards/accuracies": 0.7562500238418579,
|
| 636 |
-
"rewards/chosen": -3.8103713989257812,
|
| 637 |
-
"rewards/margins": 1.8808832168579102,
|
| 638 |
-
"rewards/rejected": -5.69125509262085,
|
| 639 |
-
"step": 440
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 1.26,
|
| 643 |
-
"learning_rate": 1.7790679834685615e-06,
|
| 644 |
-
"logits/chosen": 0.48697033524513245,
|
| 645 |
-
"logits/rejected": 0.6213686466217041,
|
| 646 |
-
"logps/chosen": -726.90185546875,
|
| 647 |
-
"logps/rejected": -848.7301635742188,
|
| 648 |
-
"loss": 0.0053,
|
| 649 |
-
"rewards/accuracies": 0.8125,
|
| 650 |
-
"rewards/chosen": -3.577058792114258,
|
| 651 |
-
"rewards/margins": 2.4176392555236816,
|
| 652 |
-
"rewards/rejected": -5.9946980476379395,
|
| 653 |
-
"step": 450
|
| 654 |
-
},
|
| 655 |
-
{
|
| 656 |
-
"epoch": 1.29,
|
| 657 |
-
"learning_rate": 1.662297450567663e-06,
|
| 658 |
-
"logits/chosen": 0.6364110708236694,
|
| 659 |
-
"logits/rejected": 0.7476580739021301,
|
| 660 |
-
"logps/chosen": -679.8997802734375,
|
| 661 |
-
"logps/rejected": -860.2080078125,
|
| 662 |
-
"loss": 0.0055,
|
| 663 |
-
"rewards/accuracies": 0.8187500238418579,
|
| 664 |
-
"rewards/chosen": -3.570373058319092,
|
| 665 |
-
"rewards/margins": 2.280672073364258,
|
| 666 |
-
"rewards/rejected": -5.85104513168335,
|
| 667 |
-
"step": 460
|
| 668 |
-
},
|
| 669 |
-
{
|
| 670 |
-
"epoch": 1.32,
|
| 671 |
-
"learning_rate": 1.5475513365797317e-06,
|
| 672 |
-
"logits/chosen": 0.47412365674972534,
|
| 673 |
-
"logits/rejected": 0.47869664430618286,
|
| 674 |
-
"logps/chosen": -706.4371337890625,
|
| 675 |
-
"logps/rejected": -842.859375,
|
| 676 |
-
"loss": 0.0054,
|
| 677 |
-
"rewards/accuracies": 0.8062499761581421,
|
| 678 |
-
"rewards/chosen": -3.590733766555786,
|
| 679 |
-
"rewards/margins": 2.3144545555114746,
|
| 680 |
-
"rewards/rejected": -5.90518856048584,
|
| 681 |
-
"step": 470
|
| 682 |
-
},
|
| 683 |
-
{
|
| 684 |
-
"epoch": 1.35,
|
| 685 |
-
"learning_rate": 1.435106940652901e-06,
|
| 686 |
-
"logits/chosen": 0.5485504865646362,
|
| 687 |
-
"logits/rejected": 0.681022584438324,
|
| 688 |
-
"logps/chosen": -682.8555908203125,
|
| 689 |
-
"logps/rejected": -810.1474609375,
|
| 690 |
-
"loss": 0.0055,
|
| 691 |
-
"rewards/accuracies": 0.800000011920929,
|
| 692 |
-
"rewards/chosen": -3.3580360412597656,
|
| 693 |
-
"rewards/margins": 2.372495174407959,
|
| 694 |
-
"rewards/rejected": -5.730531215667725,
|
| 695 |
-
"step": 480
|
| 696 |
-
},
|
| 697 |
-
{
|
| 698 |
-
"epoch": 1.38,
|
| 699 |
-
"learning_rate": 1.3252359995294376e-06,
|
| 700 |
-
"logits/chosen": 0.5862436890602112,
|
| 701 |
-
"logits/rejected": 0.6748989820480347,
|
| 702 |
-
"logps/chosen": -704.3018188476562,
|
| 703 |
-
"logps/rejected": -828.2868041992188,
|
| 704 |
-
"loss": 0.0047,
|
| 705 |
-
"rewards/accuracies": 0.8500000238418579,
|
| 706 |
-
"rewards/chosen": -3.270312786102295,
|
| 707 |
-
"rewards/margins": 2.447282075881958,
|
| 708 |
-
"rewards/rejected": -5.717595100402832,
|
| 709 |
-
"step": 490
|
| 710 |
-
},
|
| 711 |
-
{
|
| 712 |
-
"epoch": 1.4,
|
| 713 |
-
"learning_rate": 1.2182040308579942e-06,
|
| 714 |
-
"logits/chosen": 0.5599916577339172,
|
| 715 |
-
"logits/rejected": 0.6381617784500122,
|
| 716 |
-
"logps/chosen": -757.0291748046875,
|
| 717 |
-
"logps/rejected": -892.974609375,
|
| 718 |
-
"loss": 0.0041,
|
| 719 |
-
"rewards/accuracies": 0.8062499761581421,
|
| 720 |
-
"rewards/chosen": -3.7352547645568848,
|
| 721 |
-
"rewards/margins": 2.5722806453704834,
|
| 722 |
-
"rewards/rejected": -6.307535171508789,
|
| 723 |
-
"step": 500
|
| 724 |
-
},
|
| 725 |
-
{
|
| 726 |
-
"epoch": 1.43,
|
| 727 |
-
"learning_rate": 1.114269691535125e-06,
|
| 728 |
-
"logits/chosen": 0.6536546945571899,
|
| 729 |
-
"logits/rejected": 0.7505389451980591,
|
| 730 |
-
"logps/chosen": -729.1384887695312,
|
| 731 |
-
"logps/rejected": -864.0009765625,
|
| 732 |
-
"loss": 0.004,
|
| 733 |
-
"rewards/accuracies": 0.78125,
|
| 734 |
-
"rewards/chosen": -3.6905455589294434,
|
| 735 |
-
"rewards/margins": 2.2943015098571777,
|
| 736 |
-
"rewards/rejected": -5.984846591949463,
|
| 737 |
-
"step": 510
|
| 738 |
-
},
|
| 739 |
-
{
|
| 740 |
-
"epoch": 1.46,
|
| 741 |
-
"learning_rate": 1.013684152626709e-06,
|
| 742 |
-
"logits/chosen": 0.5734939575195312,
|
| 743 |
-
"logits/rejected": 0.6782180070877075,
|
| 744 |
-
"logps/chosen": -710.2030029296875,
|
| 745 |
-
"logps/rejected": -866.6712646484375,
|
| 746 |
-
"loss": 0.0042,
|
| 747 |
-
"rewards/accuracies": 0.8374999761581421,
|
| 748 |
-
"rewards/chosen": -3.7363319396972656,
|
| 749 |
-
"rewards/margins": 2.6266465187072754,
|
| 750 |
-
"rewards/rejected": -6.362977981567383,
|
| 751 |
-
"step": 520
|
| 752 |
-
},
|
| 753 |
-
{
|
| 754 |
-
"epoch": 1.49,
|
| 755 |
-
"learning_rate": 9.166904923798822e-07,
|
| 756 |
-
"logits/chosen": 0.6636993885040283,
|
| 757 |
-
"logits/rejected": 0.7554783225059509,
|
| 758 |
-
"logps/chosen": -705.7659912109375,
|
| 759 |
-
"logps/rejected": -842.1114501953125,
|
| 760 |
-
"loss": 0.0035,
|
| 761 |
-
"rewards/accuracies": 0.762499988079071,
|
| 762 |
-
"rewards/chosen": -3.721928358078003,
|
| 763 |
-
"rewards/margins": 2.2212657928466797,
|
| 764 |
-
"rewards/rejected": -5.943194389343262,
|
| 765 |
-
"step": 530
|
| 766 |
-
},
|
| 767 |
-
{
|
| 768 |
-
"epoch": 1.52,
|
| 769 |
-
"learning_rate": 8.235231087923304e-07,
|
| 770 |
-
"logits/chosen": 0.5900656580924988,
|
| 771 |
-
"logits/rejected": 0.7390304803848267,
|
| 772 |
-
"logps/chosen": -762.967041015625,
|
| 773 |
-
"logps/rejected": -907.9431762695312,
|
| 774 |
-
"loss": 0.0039,
|
| 775 |
-
"rewards/accuracies": 0.862500011920929,
|
| 776 |
-
"rewards/chosen": -3.7577643394470215,
|
| 777 |
-
"rewards/margins": 2.781780242919922,
|
| 778 |
-
"rewards/rejected": -6.539544582366943,
|
| 779 |
-
"step": 540
|
| 780 |
-
},
|
| 781 |
-
{
|
| 782 |
-
"epoch": 1.55,
|
| 783 |
-
"learning_rate": 7.344071531585462e-07,
|
| 784 |
-
"logits/chosen": 0.6735460162162781,
|
| 785 |
-
"logits/rejected": 0.7701358795166016,
|
| 786 |
-
"logps/chosen": -721.6229248046875,
|
| 787 |
-
"logps/rejected": -840.623046875,
|
| 788 |
-
"loss": 0.0035,
|
| 789 |
-
"rewards/accuracies": 0.800000011920929,
|
| 790 |
-
"rewards/chosen": -3.9318947792053223,
|
| 791 |
-
"rewards/margins": 2.4454891681671143,
|
| 792 |
-
"rewards/rejected": -6.377383232116699,
|
| 793 |
-
"step": 550
|
| 794 |
-
},
|
| 795 |
-
{
|
| 796 |
-
"epoch": 1.57,
|
| 797 |
-
"learning_rate": 6.495579859619791e-07,
|
| 798 |
-
"logits/chosen": 0.6034525632858276,
|
| 799 |
-
"logits/rejected": 0.7575315237045288,
|
| 800 |
-
"logps/chosen": -774.0142211914062,
|
| 801 |
-
"logps/rejected": -933.1998291015625,
|
| 802 |
-
"loss": 0.0038,
|
| 803 |
-
"rewards/accuracies": 0.8062499761581421,
|
| 804 |
-
"rewards/chosen": -4.213502407073975,
|
| 805 |
-
"rewards/margins": 2.61598801612854,
|
| 806 |
-
"rewards/rejected": -6.829490661621094,
|
| 807 |
-
"step": 560
|
| 808 |
-
},
|
| 809 |
-
{
|
| 810 |
-
"epoch": 1.6,
|
| 811 |
-
"learning_rate": 5.691806564279653e-07,
|
| 812 |
-
"logits/chosen": 0.6900167465209961,
|
| 813 |
-
"logits/rejected": 0.8468635678291321,
|
| 814 |
-
"logps/chosen": -746.2769775390625,
|
| 815 |
-
"logps/rejected": -889.0632934570312,
|
| 816 |
-
"loss": 0.0034,
|
| 817 |
-
"rewards/accuracies": 0.831250011920929,
|
| 818 |
-
"rewards/chosen": -3.694833278656006,
|
| 819 |
-
"rewards/margins": 2.6152822971343994,
|
| 820 |
-
"rewards/rejected": -6.310114860534668,
|
| 821 |
-
"step": 570
|
| 822 |
-
},
|
| 823 |
-
{
|
| 824 |
-
"epoch": 1.63,
|
| 825 |
-
"learning_rate": 4.934694069951748e-07,
|
| 826 |
-
"logits/chosen": 0.6488009691238403,
|
| 827 |
-
"logits/rejected": 0.6377809643745422,
|
| 828 |
-
"logps/chosen": -676.0171508789062,
|
| 829 |
-
"logps/rejected": -875.9207763671875,
|
| 830 |
-
"loss": 0.0033,
|
| 831 |
-
"rewards/accuracies": 0.8125,
|
| 832 |
-
"rewards/chosen": -3.6097030639648438,
|
| 833 |
-
"rewards/margins": 2.43593168258667,
|
| 834 |
-
"rewards/rejected": -6.04563570022583,
|
| 835 |
-
"step": 580
|
| 836 |
-
},
|
| 837 |
-
{
|
| 838 |
-
"epoch": 1.66,
|
| 839 |
-
"learning_rate": 4.22607203903091e-07,
|
| 840 |
-
"logits/chosen": 0.6412608623504639,
|
| 841 |
-
"logits/rejected": 0.7388615608215332,
|
| 842 |
-
"logps/chosen": -764.0328369140625,
|
| 843 |
-
"logps/rejected": -968.5056762695312,
|
| 844 |
-
"loss": 0.0035,
|
| 845 |
-
"rewards/accuracies": 0.856249988079071,
|
| 846 |
-
"rewards/chosen": -4.044827461242676,
|
| 847 |
-
"rewards/margins": 3.0036702156066895,
|
| 848 |
-
"rewards/rejected": -7.048497676849365,
|
| 849 |
-
"step": 590
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 1.69,
|
| 853 |
-
"learning_rate": 3.5676529502991606e-07,
|
| 854 |
-
"logits/chosen": 0.6487208604812622,
|
| 855 |
-
"logits/rejected": 0.7878293395042419,
|
| 856 |
-
"logps/chosen": -745.95556640625,
|
| 857 |
-
"logps/rejected": -838.0792236328125,
|
| 858 |
-
"loss": 0.0039,
|
| 859 |
-
"rewards/accuracies": 0.8187500238418579,
|
| 860 |
-
"rewards/chosen": -3.73905611038208,
|
| 861 |
-
"rewards/margins": 2.3786799907684326,
|
| 862 |
-
"rewards/rejected": -6.117735862731934,
|
| 863 |
-
"step": 600
|
| 864 |
-
},
|
| 865 |
-
{
|
| 866 |
-
"epoch": 1.71,
|
| 867 |
-
"learning_rate": 2.961027960494409e-07,
|
| 868 |
-
"logits/chosen": 0.6219756603240967,
|
| 869 |
-
"logits/rejected": 0.6644929647445679,
|
| 870 |
-
"logps/chosen": -760.4959106445312,
|
| 871 |
-
"logps/rejected": -936.97265625,
|
| 872 |
-
"loss": 0.0036,
|
| 873 |
-
"rewards/accuracies": 0.824999988079071,
|
| 874 |
-
"rewards/chosen": -3.9651145935058594,
|
| 875 |
-
"rewards/margins": 2.757819652557373,
|
| 876 |
-
"rewards/rejected": -6.722934722900391,
|
| 877 |
-
"step": 610
|
| 878 |
-
},
|
| 879 |
-
{
|
| 880 |
-
"epoch": 1.74,
|
| 881 |
-
"learning_rate": 2.4076630590699065e-07,
|
| 882 |
-
"logits/chosen": 0.6320921182632446,
|
| 883 |
-
"logits/rejected": 0.7059046626091003,
|
| 884 |
-
"logps/chosen": -700.6550903320312,
|
| 885 |
-
"logps/rejected": -826.9192504882812,
|
| 886 |
-
"loss": 0.0037,
|
| 887 |
-
"rewards/accuracies": 0.8500000238418579,
|
| 888 |
-
"rewards/chosen": -3.5051655769348145,
|
| 889 |
-
"rewards/margins": 2.5177574157714844,
|
| 890 |
-
"rewards/rejected": -6.022922515869141,
|
| 891 |
-
"step": 620
|
| 892 |
-
},
|
| 893 |
-
{
|
| 894 |
-
"epoch": 1.77,
|
| 895 |
-
"learning_rate": 1.9088955254370351e-07,
|
| 896 |
-
"logits/chosen": 0.5824363827705383,
|
| 897 |
-
"logits/rejected": 0.765023946762085,
|
| 898 |
-
"logps/chosen": -715.6721801757812,
|
| 899 |
-
"logps/rejected": -855.0099487304688,
|
| 900 |
-
"loss": 0.0042,
|
| 901 |
-
"rewards/accuracies": 0.793749988079071,
|
| 902 |
-
"rewards/chosen": -3.933236598968506,
|
| 903 |
-
"rewards/margins": 2.2994723320007324,
|
| 904 |
-
"rewards/rejected": -6.232708930969238,
|
| 905 |
-
"step": 630
|
| 906 |
-
},
|
| 907 |
-
{
|
| 908 |
-
"epoch": 1.8,
|
| 909 |
-
"learning_rate": 1.465930697252882e-07,
|
| 910 |
-
"logits/chosen": 0.5748051404953003,
|
| 911 |
-
"logits/rejected": 0.668735146522522,
|
| 912 |
-
"logps/chosen": -811.7765502929688,
|
| 913 |
-
"logps/rejected": -966.9187622070312,
|
| 914 |
-
"loss": 0.0032,
|
| 915 |
-
"rewards/accuracies": 0.862500011920929,
|
| 916 |
-
"rewards/chosen": -4.106084823608398,
|
| 917 |
-
"rewards/margins": 2.9647626876831055,
|
| 918 |
-
"rewards/rejected": -7.070847988128662,
|
| 919 |
-
"step": 640
|
| 920 |
-
},
|
| 921 |
-
{
|
| 922 |
-
"epoch": 1.83,
|
| 923 |
-
"learning_rate": 1.0798390575624762e-07,
|
| 924 |
-
"logits/chosen": 0.7111840844154358,
|
| 925 |
-
"logits/rejected": 0.8005073666572571,
|
| 926 |
-
"logps/chosen": -721.6022338867188,
|
| 927 |
-
"logps/rejected": -854.2897338867188,
|
| 928 |
-
"loss": 0.0033,
|
| 929 |
-
"rewards/accuracies": 0.8374999761581421,
|
| 930 |
-
"rewards/chosen": -3.6958014965057373,
|
| 931 |
-
"rewards/margins": 2.6302475929260254,
|
| 932 |
-
"rewards/rejected": -6.326048851013184,
|
| 933 |
-
"step": 650
|
| 934 |
-
},
|
| 935 |
-
{
|
| 936 |
-
"epoch": 1.85,
|
| 937 |
-
"learning_rate": 7.515536478350849e-08,
|
| 938 |
-
"logits/chosen": 0.6084551215171814,
|
| 939 |
-
"logits/rejected": 0.6986690163612366,
|
| 940 |
-
"logps/chosen": -774.9620361328125,
|
| 941 |
-
"logps/rejected": -953.1961669921875,
|
| 942 |
-
"loss": 0.0036,
|
| 943 |
-
"rewards/accuracies": 0.875,
|
| 944 |
-
"rewards/chosen": -3.89312481880188,
|
| 945 |
-
"rewards/margins": 2.904383420944214,
|
| 946 |
-
"rewards/rejected": -6.79750919342041,
|
| 947 |
-
"step": 660
|
| 948 |
-
},
|
| 949 |
-
{
|
| 950 |
-
"epoch": 1.88,
|
| 951 |
-
"learning_rate": 4.818678131461862e-08,
|
| 952 |
-
"logits/chosen": 0.7103031277656555,
|
| 953 |
-
"logits/rejected": 0.8091765642166138,
|
| 954 |
-
"logps/chosen": -720.9351806640625,
|
| 955 |
-
"logps/rejected": -879.3292846679688,
|
| 956 |
-
"loss": 0.0028,
|
| 957 |
-
"rewards/accuracies": 0.8374999761581421,
|
| 958 |
-
"rewards/chosen": -3.853041887283325,
|
| 959 |
-
"rewards/margins": 2.506618022918701,
|
| 960 |
-
"rewards/rejected": -6.359659671783447,
|
| 961 |
-
"step": 670
|
| 962 |
-
},
|
| 963 |
-
{
|
| 964 |
-
"epoch": 1.91,
|
| 965 |
-
"learning_rate": 2.7143328495423948e-08,
|
| 966 |
-
"logits/chosen": 0.5255075693130493,
|
| 967 |
-
"logits/rejected": 0.7452009320259094,
|
| 968 |
-
"logps/chosen": -735.2042236328125,
|
| 969 |
-
"logps/rejected": -863.3488159179688,
|
| 970 |
-
"loss": 0.0034,
|
| 971 |
-
"rewards/accuracies": 0.824999988079071,
|
| 972 |
-
"rewards/chosen": -3.803483247756958,
|
| 973 |
-
"rewards/margins": 2.5907998085021973,
|
| 974 |
-
"rewards/rejected": -6.394282817840576,
|
| 975 |
-
"step": 680
|
| 976 |
-
},
|
| 977 |
-
{
|
| 978 |
-
"epoch": 1.94,
|
| 979 |
-
"learning_rate": 1.2075860610551715e-08,
|
| 980 |
-
"logits/chosen": 0.4812963604927063,
|
| 981 |
-
"logits/rejected": 0.5726746320724487,
|
| 982 |
-
"logps/chosen": -796.3458251953125,
|
| 983 |
-
"logps/rejected": -932.7238159179688,
|
| 984 |
-
"loss": 0.0033,
|
| 985 |
-
"rewards/accuracies": 0.8374999761581421,
|
| 986 |
-
"rewards/chosen": -4.022668361663818,
|
| 987 |
-
"rewards/margins": 2.6870803833007812,
|
| 988 |
-
"rewards/rejected": -6.7097487449646,
|
| 989 |
-
"step": 690
|
| 990 |
-
},
|
| 991 |
-
{
|
| 992 |
-
"epoch": 1.97,
|
| 993 |
-
"learning_rate": 3.0207901873091638e-09,
|
| 994 |
-
"logits/chosen": 0.6011410355567932,
|
| 995 |
-
"logits/rejected": 0.7141892313957214,
|
| 996 |
-
"logps/chosen": -788.8591918945312,
|
| 997 |
-
"logps/rejected": -992.2294921875,
|
| 998 |
-
"loss": 0.0025,
|
| 999 |
-
"rewards/accuracies": 0.831250011920929,
|
| 1000 |
-
"rewards/chosen": -4.254032611846924,
|
| 1001 |
-
"rewards/margins": 2.852738380432129,
|
| 1002 |
-
"rewards/rejected": -7.1067705154418945,
|
| 1003 |
-
"step": 700
|
| 1004 |
-
},
|
| 1005 |
-
{
|
| 1006 |
-
"epoch": 2.0,
|
| 1007 |
-
"learning_rate": 0.0,
|
| 1008 |
-
"logits/chosen": 0.6989771127700806,
|
| 1009 |
-
"logits/rejected": 0.749651312828064,
|
| 1010 |
-
"logps/chosen": -713.0648803710938,
|
| 1011 |
-
"logps/rejected": -953.052734375,
|
| 1012 |
-
"loss": 0.0041,
|
| 1013 |
-
"rewards/accuracies": 0.8500000238418579,
|
| 1014 |
-
"rewards/chosen": -3.8918204307556152,
|
| 1015 |
-
"rewards/margins": 3.1703882217407227,
|
| 1016 |
-
"rewards/rejected": -7.062208652496338,
|
| 1017 |
-
"step": 710
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
-
"epoch":
|
| 1021 |
-
"step":
|
| 1022 |
"total_flos": 0.0,
|
| 1023 |
-
"train_loss": 0.
|
| 1024 |
-
"train_runtime":
|
| 1025 |
-
"train_samples_per_second":
|
| 1026 |
-
"train_steps_per_second": 0.
|
| 1027 |
}
|
| 1028 |
],
|
| 1029 |
"logging_steps": 10,
|
| 1030 |
-
"max_steps":
|
| 1031 |
-
"num_train_epochs":
|
| 1032 |
"save_steps": 10000,
|
| 1033 |
"total_flos": 0.0,
|
| 1034 |
"trial_name": null,
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9975412715138743,
|
| 5 |
"eval_steps": 10000,
|
| 6 |
+
"global_step": 355,
|
| 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": 2.7777777777777774e-08,
|
| 14 |
+
"logits/chosen": -0.13174405694007874,
|
| 15 |
+
"logits/rejected": -0.027169257402420044,
|
| 16 |
+
"logps/chosen": -477.4691162109375,
|
| 17 |
+
"logps/rejected": -277.6482238769531,
|
| 18 |
+
"loss": 0.4106,
|
| 19 |
"rewards/accuracies": 0.0,
|
| 20 |
"rewards/chosen": 0.0,
|
| 21 |
"rewards/margins": 0.0,
|
|
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.03,
|
| 27 |
+
"learning_rate": 2.7777777777777776e-07,
|
| 28 |
+
"logits/chosen": -0.0896572694182396,
|
| 29 |
+
"logits/rejected": -0.04708625003695488,
|
| 30 |
+
"logps/chosen": -334.1234130859375,
|
| 31 |
+
"logps/rejected": -264.19927978515625,
|
| 32 |
+
"loss": 0.4187,
|
| 33 |
+
"rewards/accuracies": 0.4027777910232544,
|
| 34 |
+
"rewards/chosen": -0.0011318529723212123,
|
| 35 |
+
"rewards/margins": -0.0006679879734292626,
|
| 36 |
+
"rewards/rejected": -0.0004638649697881192,
|
| 37 |
"step": 10
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.06,
|
| 41 |
+
"learning_rate": 5.555555555555555e-07,
|
| 42 |
+
"logits/chosen": -0.06541652977466583,
|
| 43 |
+
"logits/rejected": -0.027149802073836327,
|
| 44 |
+
"logps/chosen": -312.1936950683594,
|
| 45 |
+
"logps/rejected": -212.1822967529297,
|
| 46 |
+
"loss": 0.422,
|
| 47 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 48 |
+
"rewards/chosen": 0.0018345726421102881,
|
| 49 |
+
"rewards/margins": 0.008850323967635632,
|
| 50 |
+
"rewards/rejected": -0.007015751209110022,
|
| 51 |
"step": 20
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.08,
|
| 55 |
+
"learning_rate": 8.333333333333333e-07,
|
| 56 |
+
"logits/chosen": -0.033993594348430634,
|
| 57 |
+
"logits/rejected": 0.014452556148171425,
|
| 58 |
+
"logps/chosen": -369.52886962890625,
|
| 59 |
+
"logps/rejected": -227.0442657470703,
|
| 60 |
+
"loss": 0.4287,
|
| 61 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 62 |
+
"rewards/chosen": 0.012805985286831856,
|
| 63 |
+
"rewards/margins": 0.05532551556825638,
|
| 64 |
+
"rewards/rejected": -0.04251953214406967,
|
| 65 |
"step": 30
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.11,
|
| 69 |
+
"learning_rate": 9.99612097830993e-07,
|
| 70 |
+
"logits/chosen": -0.049494121223688126,
|
| 71 |
+
"logits/rejected": -0.007341804448515177,
|
| 72 |
+
"logps/chosen": -328.2823791503906,
|
| 73 |
+
"logps/rejected": -251.8525848388672,
|
| 74 |
+
"loss": 0.4553,
|
| 75 |
"rewards/accuracies": 0.5687500238418579,
|
| 76 |
+
"rewards/chosen": -0.03969588130712509,
|
| 77 |
+
"rewards/margins": 0.06737084686756134,
|
| 78 |
+
"rewards/rejected": -0.10706672817468643,
|
| 79 |
"step": 40
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.14,
|
| 83 |
+
"learning_rate": 9.952551076085863e-07,
|
| 84 |
+
"logits/chosen": -0.054784227162599564,
|
| 85 |
+
"logits/rejected": -0.018202614039182663,
|
| 86 |
+
"logps/chosen": -343.4543762207031,
|
| 87 |
+
"logps/rejected": -278.887451171875,
|
| 88 |
+
"loss": 0.478,
|
| 89 |
+
"rewards/accuracies": 0.606249988079071,
|
| 90 |
+
"rewards/chosen": -0.029784226790070534,
|
| 91 |
+
"rewards/margins": 0.14676395058631897,
|
| 92 |
+
"rewards/rejected": -0.17654818296432495,
|
| 93 |
"step": 50
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.17,
|
| 97 |
+
"learning_rate": 9.860986139994238e-07,
|
| 98 |
+
"logits/chosen": -0.17503580451011658,
|
| 99 |
+
"logits/rejected": -0.10935833305120468,
|
| 100 |
+
"logps/chosen": -399.97161865234375,
|
| 101 |
+
"logps/rejected": -245.5420684814453,
|
| 102 |
+
"loss": 0.4856,
|
| 103 |
+
"rewards/accuracies": 0.6875,
|
| 104 |
+
"rewards/chosen": 0.0788765698671341,
|
| 105 |
+
"rewards/margins": 0.30753207206726074,
|
| 106 |
+
"rewards/rejected": -0.22865548729896545,
|
| 107 |
"step": 60
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.2,
|
| 111 |
+
"learning_rate": 9.722313523268027e-07,
|
| 112 |
+
"logits/chosen": -0.13078172504901886,
|
| 113 |
+
"logits/rejected": -0.018874743953347206,
|
| 114 |
+
"logps/chosen": -382.87396240234375,
|
| 115 |
+
"logps/rejected": -252.6072540283203,
|
| 116 |
+
"loss": 0.4667,
|
| 117 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 118 |
+
"rewards/chosen": 0.09564249962568283,
|
| 119 |
+
"rewards/margins": 0.21217863261699677,
|
| 120 |
+
"rewards/rejected": -0.11653614044189453,
|
| 121 |
"step": 70
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.22,
|
| 125 |
+
"learning_rate": 9.537877098354784e-07,
|
| 126 |
+
"logits/chosen": 0.019111448898911476,
|
| 127 |
+
"logits/rejected": 0.04028189927339554,
|
| 128 |
+
"logps/chosen": -277.33154296875,
|
| 129 |
+
"logps/rejected": -215.5694122314453,
|
| 130 |
+
"loss": 0.4657,
|
| 131 |
+
"rewards/accuracies": 0.606249988079071,
|
| 132 |
+
"rewards/chosen": 0.01853206194937229,
|
| 133 |
+
"rewards/margins": 0.18713198602199554,
|
| 134 |
+
"rewards/rejected": -0.168599933385849,
|
| 135 |
"step": 80
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.25,
|
| 139 |
+
"learning_rate": 9.309464233486386e-07,
|
| 140 |
+
"logits/chosen": -0.184749573469162,
|
| 141 |
+
"logits/rejected": -0.12197474390268326,
|
| 142 |
+
"logps/chosen": -374.56268310546875,
|
| 143 |
+
"logps/rejected": -224.47860717773438,
|
| 144 |
+
"loss": 0.4724,
|
| 145 |
"rewards/accuracies": 0.71875,
|
| 146 |
+
"rewards/chosen": 0.19164375960826874,
|
| 147 |
+
"rewards/margins": 0.33850011229515076,
|
| 148 |
+
"rewards/rejected": -0.1468563675880432,
|
| 149 |
"step": 90
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.28,
|
| 153 |
+
"learning_rate": 9.039288471343504e-07,
|
| 154 |
+
"logits/chosen": -0.06358620524406433,
|
| 155 |
+
"logits/rejected": -0.022323714569211006,
|
| 156 |
+
"logps/chosen": -352.3625183105469,
|
| 157 |
+
"logps/rejected": -265.12457275390625,
|
| 158 |
+
"loss": 0.4579,
|
| 159 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 160 |
+
"rewards/chosen": 0.11637835204601288,
|
| 161 |
+
"rewards/margins": 0.15370506048202515,
|
| 162 |
+
"rewards/rejected": -0.03732669726014137,
|
| 163 |
"step": 100
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.31,
|
| 167 |
+
"learning_rate": 8.729968077675454e-07,
|
| 168 |
+
"logits/chosen": -0.16022691130638123,
|
| 169 |
+
"logits/rejected": -0.06551636755466461,
|
| 170 |
+
"logps/chosen": -304.0919189453125,
|
| 171 |
+
"logps/rejected": -257.5033874511719,
|
| 172 |
+
"loss": 0.4444,
|
| 173 |
+
"rewards/accuracies": 0.606249988079071,
|
| 174 |
+
"rewards/chosen": 0.09323982149362564,
|
| 175 |
+
"rewards/margins": 0.10607878863811493,
|
| 176 |
+
"rewards/rejected": -0.012838983908295631,
|
| 177 |
"step": 110
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.34,
|
| 181 |
+
"learning_rate": 8.384500667760089e-07,
|
| 182 |
+
"logits/chosen": -0.18294575810432434,
|
| 183 |
+
"logits/rejected": -0.1334661990404129,
|
| 184 |
+
"logps/chosen": -323.1263427734375,
|
| 185 |
+
"logps/rejected": -206.44387817382812,
|
| 186 |
+
"loss": 0.4535,
|
| 187 |
+
"rewards/accuracies": 0.65625,
|
| 188 |
+
"rewards/chosen": 0.17489886283874512,
|
| 189 |
+
"rewards/margins": 0.1883935183286667,
|
| 190 |
+
"rewards/rejected": -0.013494668528437614,
|
| 191 |
"step": 120
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.37,
|
| 195 |
+
"learning_rate": 8.006234156598042e-07,
|
| 196 |
+
"logits/chosen": -0.09687581658363342,
|
| 197 |
+
"logits/rejected": -0.0031311712227761745,
|
| 198 |
+
"logps/chosen": -361.1056823730469,
|
| 199 |
+
"logps/rejected": -219.41552734375,
|
| 200 |
+
"loss": 0.4484,
|
| 201 |
+
"rewards/accuracies": 0.625,
|
| 202 |
+
"rewards/chosen": 0.12578140199184418,
|
| 203 |
+
"rewards/margins": 0.2744066119194031,
|
| 204 |
+
"rewards/rejected": -0.1486252248287201,
|
| 205 |
"step": 130
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.39,
|
| 209 |
+
"learning_rate": 7.59883431436215e-07,
|
| 210 |
+
"logits/chosen": -0.03516136482357979,
|
| 211 |
+
"logits/rejected": -0.005446717143058777,
|
| 212 |
+
"logps/chosen": -316.314208984375,
|
| 213 |
+
"logps/rejected": -241.97024536132812,
|
| 214 |
+
"loss": 0.4383,
|
| 215 |
+
"rewards/accuracies": 0.625,
|
| 216 |
+
"rewards/chosen": 0.025916021317243576,
|
| 217 |
+
"rewards/margins": 0.16546496748924255,
|
| 218 |
+
"rewards/rejected": -0.13954894244670868,
|
| 219 |
"step": 140
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.42,
|
| 223 |
+
"learning_rate": 7.166249241521318e-07,
|
| 224 |
+
"logits/chosen": 0.0030886970926076174,
|
| 225 |
+
"logits/rejected": 0.06723493337631226,
|
| 226 |
+
"logps/chosen": -293.86627197265625,
|
| 227 |
+
"logps/rejected": -255.26492309570312,
|
| 228 |
+
"loss": 0.4382,
|
| 229 |
+
"rewards/accuracies": 0.606249988079071,
|
| 230 |
+
"rewards/chosen": -0.05645722150802612,
|
| 231 |
+
"rewards/margins": 0.17538480460643768,
|
| 232 |
+
"rewards/rejected": -0.2318420112133026,
|
| 233 |
"step": 150
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.45,
|
| 237 |
+
"learning_rate": 6.712671107909358e-07,
|
| 238 |
+
"logits/chosen": -0.03268152475357056,
|
| 239 |
+
"logits/rejected": 0.12709534168243408,
|
| 240 |
+
"logps/chosen": -369.74859619140625,
|
| 241 |
+
"logps/rejected": -260.13128662109375,
|
| 242 |
+
"loss": 0.4255,
|
| 243 |
+
"rewards/accuracies": 0.643750011920929,
|
| 244 |
+
"rewards/chosen": 0.003464625682681799,
|
| 245 |
+
"rewards/margins": 0.2506219744682312,
|
| 246 |
+
"rewards/rejected": -0.24715733528137207,
|
| 247 |
"step": 160
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.48,
|
| 251 |
+
"learning_rate": 6.24249552652447e-07,
|
| 252 |
+
"logits/chosen": 0.04178273305296898,
|
| 253 |
+
"logits/rejected": 0.12335582822561264,
|
| 254 |
+
"logps/chosen": -316.4767150878906,
|
| 255 |
+
"logps/rejected": -268.18829345703125,
|
| 256 |
+
"loss": 0.4169,
|
| 257 |
+
"rewards/accuracies": 0.59375,
|
| 258 |
+
"rewards/chosen": -0.006209957879036665,
|
| 259 |
+
"rewards/margins": 0.20326288044452667,
|
| 260 |
+
"rewards/rejected": -0.20947282016277313,
|
| 261 |
"step": 170
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.51,
|
| 265 |
+
"learning_rate": 5.760278955766694e-07,
|
| 266 |
+
"logits/chosen": -0.12427058070898056,
|
| 267 |
+
"logits/rejected": 0.009830540046095848,
|
| 268 |
+
"logps/chosen": -327.13958740234375,
|
| 269 |
+
"logps/rejected": -258.3717041015625,
|
| 270 |
+
"loss": 0.4267,
|
| 271 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 272 |
+
"rewards/chosen": -0.042253121733665466,
|
| 273 |
+
"rewards/margins": 0.19665148854255676,
|
| 274 |
+
"rewards/rejected": -0.23890459537506104,
|
| 275 |
"step": 180
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.53,
|
| 279 |
+
"learning_rate": 5.270694542927088e-07,
|
| 280 |
+
"logits/chosen": -0.16560761630535126,
|
| 281 |
+
"logits/rejected": -0.04073227569460869,
|
| 282 |
+
"logps/chosen": -341.1844482421875,
|
| 283 |
+
"logps/rejected": -227.2850341796875,
|
| 284 |
+
"loss": 0.4261,
|
| 285 |
"rewards/accuracies": 0.668749988079071,
|
| 286 |
+
"rewards/chosen": 0.0071119763888418674,
|
| 287 |
+
"rewards/margins": 0.2952454090118408,
|
| 288 |
+
"rewards/rejected": -0.288133442401886,
|
| 289 |
"step": 190
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.56,
|
| 293 |
+
"learning_rate": 4.778486836848107e-07,
|
| 294 |
+
"logits/chosen": -0.007979141548275948,
|
| 295 |
+
"logits/rejected": 0.1243690699338913,
|
| 296 |
+
"logps/chosen": -329.4273986816406,
|
| 297 |
+
"logps/rejected": -260.57806396484375,
|
| 298 |
+
"loss": 0.4096,
|
| 299 |
+
"rewards/accuracies": 0.5625,
|
| 300 |
+
"rewards/chosen": -0.19496320188045502,
|
| 301 |
+
"rewards/margins": 0.16824397444725037,
|
| 302 |
+
"rewards/rejected": -0.3632071614265442,
|
| 303 |
"step": 200
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.59,
|
| 307 |
+
"learning_rate": 4.2884258086335745e-07,
|
| 308 |
+
"logits/chosen": 0.09336410462856293,
|
| 309 |
+
"logits/rejected": 0.19506987929344177,
|
| 310 |
+
"logps/chosen": -391.4615173339844,
|
| 311 |
+
"logps/rejected": -279.2521057128906,
|
| 312 |
+
"loss": 0.4003,
|
| 313 |
+
"rewards/accuracies": 0.59375,
|
| 314 |
+
"rewards/chosen": -0.11562051624059677,
|
| 315 |
+
"rewards/margins": 0.245724156498909,
|
| 316 |
+
"rewards/rejected": -0.36134466528892517,
|
| 317 |
"step": 210
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.62,
|
| 321 |
+
"learning_rate": 3.8052606259922095e-07,
|
| 322 |
+
"logits/chosen": -0.16688141226768494,
|
| 323 |
+
"logits/rejected": -0.08500812947750092,
|
| 324 |
+
"logps/chosen": -362.3302917480469,
|
| 325 |
+
"logps/rejected": -247.5942840576172,
|
| 326 |
+
"loss": 0.4244,
|
| 327 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 328 |
+
"rewards/chosen": -0.07475811243057251,
|
| 329 |
+
"rewards/margins": 0.20620782673358917,
|
| 330 |
+
"rewards/rejected": -0.28096598386764526,
|
| 331 |
"step": 220
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.65,
|
| 335 |
+
"learning_rate": 3.333673629186279e-07,
|
| 336 |
+
"logits/chosen": -0.02717510424554348,
|
| 337 |
+
"logits/rejected": 0.12363864481449127,
|
| 338 |
+
"logps/chosen": -332.33319091796875,
|
| 339 |
+
"logps/rejected": -247.22817993164062,
|
| 340 |
+
"loss": 0.4115,
|
| 341 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 342 |
+
"rewards/chosen": -0.08057795464992523,
|
| 343 |
+
"rewards/margins": 0.2381751984357834,
|
| 344 |
+
"rewards/rejected": -0.3187531530857086,
|
| 345 |
"step": 230
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.67,
|
| 349 |
+
"learning_rate": 2.878234954603167e-07,
|
| 350 |
+
"logits/chosen": 0.033598482608795166,
|
| 351 |
+
"logits/rejected": 0.18793973326683044,
|
| 352 |
+
"logps/chosen": -381.00506591796875,
|
| 353 |
+
"logps/rejected": -270.8756103515625,
|
| 354 |
+
"loss": 0.3798,
|
| 355 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 356 |
+
"rewards/chosen": -0.12533587217330933,
|
| 357 |
+
"rewards/margins": 0.24204333126544952,
|
| 358 |
+
"rewards/rejected": -0.36737921833992004,
|
| 359 |
"step": 240
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.7,
|
| 363 |
+
"learning_rate": 2.443358245691555e-07,
|
| 364 |
+
"logits/chosen": 0.030673842877149582,
|
| 365 |
+
"logits/rejected": 0.18992134928703308,
|
| 366 |
+
"logps/chosen": -383.7073059082031,
|
| 367 |
+
"logps/rejected": -261.9964294433594,
|
| 368 |
+
"loss": 0.3877,
|
| 369 |
"rewards/accuracies": 0.6625000238418579,
|
| 370 |
+
"rewards/chosen": -0.09187673032283783,
|
| 371 |
+
"rewards/margins": 0.3298332989215851,
|
| 372 |
+
"rewards/rejected": -0.4217100143432617,
|
| 373 |
"step": 250
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.73,
|
| 377 |
+
"learning_rate": 2.0332578804662782e-07,
|
| 378 |
+
"logits/chosen": 0.024305405095219612,
|
| 379 |
+
"logits/rejected": 0.132650688290596,
|
| 380 |
+
"logps/chosen": -368.91131591796875,
|
| 381 |
+
"logps/rejected": -269.9962463378906,
|
| 382 |
+
"loss": 0.4026,
|
| 383 |
"rewards/accuracies": 0.625,
|
| 384 |
+
"rewards/chosen": -0.14426225423812866,
|
| 385 |
+
"rewards/margins": 0.2727457880973816,
|
| 386 |
+
"rewards/rejected": -0.41700801253318787,
|
| 387 |
"step": 260
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.76,
|
| 391 |
+
"learning_rate": 1.651908130088947e-07,
|
| 392 |
+
"logits/chosen": 0.13495397567749023,
|
| 393 |
+
"logits/rejected": 0.21630129218101501,
|
| 394 |
+
"logps/chosen": -346.6638488769531,
|
| 395 |
+
"logps/rejected": -274.6488952636719,
|
| 396 |
+
"loss": 0.3821,
|
| 397 |
+
"rewards/accuracies": 0.668749988079071,
|
| 398 |
+
"rewards/chosen": -0.25730255246162415,
|
| 399 |
+
"rewards/margins": 0.2660498023033142,
|
| 400 |
+
"rewards/rejected": -0.5233522653579712,
|
| 401 |
"step": 270
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.79,
|
| 405 |
+
"learning_rate": 1.3030046443173442e-07,
|
| 406 |
+
"logits/chosen": 0.12753400206565857,
|
| 407 |
+
"logits/rejected": 0.26089444756507874,
|
| 408 |
+
"logps/chosen": -396.9707336425781,
|
| 409 |
+
"logps/rejected": -273.0684814453125,
|
| 410 |
+
"loss": 0.4015,
|
| 411 |
+
"rewards/accuracies": 0.625,
|
| 412 |
+
"rewards/chosen": -0.19528909027576447,
|
| 413 |
+
"rewards/margins": 0.3000025451183319,
|
| 414 |
+
"rewards/rejected": -0.49529165029525757,
|
| 415 |
"step": 280
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 0.81,
|
| 419 |
+
"learning_rate": 9.899286370670574e-08,
|
| 420 |
+
"logits/chosen": 0.18344645202159882,
|
| 421 |
+
"logits/rejected": 0.3353565037250519,
|
| 422 |
+
"logps/chosen": -358.2286376953125,
|
| 423 |
+
"logps/rejected": -288.5234680175781,
|
| 424 |
+
"loss": 0.4025,
|
| 425 |
+
"rewards/accuracies": 0.5687500238418579,
|
| 426 |
+
"rewards/chosen": -0.28220534324645996,
|
| 427 |
+
"rewards/margins": 0.19654087722301483,
|
| 428 |
+
"rewards/rejected": -0.478746235370636,
|
| 429 |
"step": 290
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.84,
|
| 433 |
+
"learning_rate": 7.157141191620548e-08,
|
| 434 |
+
"logits/chosen": 0.0641961470246315,
|
| 435 |
+
"logits/rejected": 0.2366667091846466,
|
| 436 |
+
"logps/chosen": -380.06103515625,
|
| 437 |
+
"logps/rejected": -267.75213623046875,
|
| 438 |
+
"loss": 0.3997,
|
| 439 |
+
"rewards/accuracies": 0.625,
|
| 440 |
+
"rewards/chosen": -0.13748347759246826,
|
| 441 |
+
"rewards/margins": 0.3018878996372223,
|
| 442 |
+
"rewards/rejected": -0.43937140703201294,
|
| 443 |
"step": 300
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 0.87,
|
| 447 |
+
"learning_rate": 4.830184958207006e-08,
|
| 448 |
+
"logits/chosen": 0.03403336927294731,
|
| 449 |
+
"logits/rejected": 0.19396355748176575,
|
| 450 |
+
"logps/chosen": -347.7532653808594,
|
| 451 |
+
"logps/rejected": -284.3299865722656,
|
| 452 |
+
"loss": 0.4026,
|
| 453 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 454 |
+
"rewards/chosen": -0.16916589438915253,
|
| 455 |
+
"rewards/margins": 0.25709637999534607,
|
| 456 |
+
"rewards/rejected": -0.4262623190879822,
|
| 457 |
"step": 310
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.9,
|
| 461 |
+
"learning_rate": 2.940968138161731e-08,
|
| 462 |
+
"logits/chosen": 0.11429999023675919,
|
| 463 |
+
"logits/rejected": 0.17834721505641937,
|
| 464 |
+
"logps/chosen": -330.48284912109375,
|
| 465 |
+
"logps/rejected": -274.64727783203125,
|
| 466 |
+
"loss": 0.3982,
|
| 467 |
+
"rewards/accuracies": 0.612500011920929,
|
| 468 |
+
"rewards/chosen": -0.1756475865840912,
|
| 469 |
+
"rewards/margins": 0.1860581338405609,
|
| 470 |
+
"rewards/rejected": -0.3617057204246521,
|
| 471 |
"step": 320
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 0.93,
|
| 475 |
+
"learning_rate": 1.507799078812799e-08,
|
| 476 |
+
"logits/chosen": -0.007492154836654663,
|
| 477 |
+
"logits/rejected": 0.10714348405599594,
|
| 478 |
+
"logps/chosen": -425.7674865722656,
|
| 479 |
+
"logps/rejected": -314.00860595703125,
|
| 480 |
+
"loss": 0.3971,
|
| 481 |
+
"rewards/accuracies": 0.6187499761581421,
|
| 482 |
+
"rewards/chosen": -0.17962414026260376,
|
| 483 |
+
"rewards/margins": 0.292041152715683,
|
| 484 |
+
"rewards/rejected": -0.47166532278060913,
|
| 485 |
"step": 330
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"epoch": 0.96,
|
| 489 |
+
"learning_rate": 5.445665814031941e-09,
|
| 490 |
+
"logits/chosen": 0.06636445224285126,
|
| 491 |
+
"logits/rejected": 0.1753680408000946,
|
| 492 |
+
"logps/chosen": -369.9612731933594,
|
| 493 |
+
"logps/rejected": -278.1523132324219,
|
| 494 |
+
"loss": 0.4052,
|
| 495 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 496 |
+
"rewards/chosen": -0.18568792939186096,
|
| 497 |
+
"rewards/margins": 0.26998209953308105,
|
| 498 |
+
"rewards/rejected": -0.4556700587272644,
|
| 499 |
"step": 340
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 0.98,
|
| 503 |
+
"learning_rate": 6.060530510659245e-10,
|
| 504 |
+
"logits/chosen": 0.03166942670941353,
|
| 505 |
+
"logits/rejected": 0.10389814525842667,
|
| 506 |
+
"logps/chosen": -357.81988525390625,
|
| 507 |
+
"logps/rejected": -275.1876525878906,
|
| 508 |
+
"loss": 0.4038,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 509 |
"rewards/accuracies": 0.6625000238418579,
|
| 510 |
+
"rewards/chosen": -0.20625650882720947,
|
| 511 |
+
"rewards/margins": 0.26138800382614136,
|
| 512 |
+
"rewards/rejected": -0.46764451265335083,
|
| 513 |
+
"step": 350
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 1.0,
|
| 517 |
+
"step": 355,
|
| 518 |
"total_flos": 0.0,
|
| 519 |
+
"train_loss": 0.4252443082735572,
|
| 520 |
+
"train_runtime": 5295.2592,
|
| 521 |
+
"train_samples_per_second": 8.602,
|
| 522 |
+
"train_steps_per_second": 0.067
|
| 523 |
}
|
| 524 |
],
|
| 525 |
"logging_steps": 10,
|
| 526 |
+
"max_steps": 355,
|
| 527 |
+
"num_train_epochs": 1,
|
| 528 |
"save_steps": 10000,
|
| 529 |
"total_flos": 0.0,
|
| 530 |
"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 sha256:02106212d8646cd330ea93c455fab2db74b591fce43575560426efd71655c54f
|
| 3 |
size 6008
|