Model save
Browse files- README.md +1 -1
- all_results.json +3 -3
- 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 +3 -3
- trainer_state.json +384 -384
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -27,7 +27,7 @@ print(output["generated_text"])
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/gongc1995-city-university-of-hong-kong/huggingface/runs/
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
|
|
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/gongc1995-city-university-of-hong-kong/huggingface/runs/299z2w0y)
|
| 31 |
|
| 32 |
|
| 33 |
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 7500,
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
"train_steps_per_second": 0.006
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 2.3046215271270145,
|
| 4 |
+
"train_runtime": 39674.2598,
|
| 5 |
"train_samples": 7500,
|
| 6 |
+
"train_samples_per_second": 0.756,
|
| 7 |
"train_steps_per_second": 0.006
|
| 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 4877660776
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:854567b961c7917403b178e4175694cad2584d65934dc4a5ab8586789d620207
|
| 3 |
size 4877660776
|
model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4932751008
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd4d1eece6c6dda70828314d370c5ff174f5125da6bb7145eac618deb24560fb
|
| 3 |
size 4932751008
|
model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4330865200
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81be49c5316cad8f273d0e521fad4c2096739044571d0c5a4ad20f9c890408ff
|
| 3 |
size 4330865200
|
model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1089994880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68a49f902ce8187358fdd05fdd999a321034f8694dc6291a93dbad48c58660ff
|
| 3 |
size 1089994880
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 7500,
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
"train_steps_per_second": 0.006
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 2.3046215271270145,
|
| 4 |
+
"train_runtime": 39674.2598,
|
| 5 |
"train_samples": 7500,
|
| 6 |
+
"train_samples_per_second": 0.756,
|
| 7 |
"train_steps_per_second": 0.006
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -12,7 +12,7 @@
|
|
| 12 |
"clip_ratio": 0.0,
|
| 13 |
"completion_length": 445.9810485839844,
|
| 14 |
"epoch": 0.017057569296375266,
|
| 15 |
-
"grad_norm": 0.
|
| 16 |
"kl": 0.0,
|
| 17 |
"learning_rate": 1.25e-07,
|
| 18 |
"loss": -0.0061,
|
|
@@ -24,678 +24,678 @@
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"clip_ratio": 0.0,
|
| 27 |
-
"completion_length":
|
| 28 |
"epoch": 0.08528784648187633,
|
| 29 |
-
"grad_norm": 0.
|
| 30 |
-
"kl":
|
| 31 |
"learning_rate": 6.25e-07,
|
| 32 |
-
"loss": -0.
|
| 33 |
-
"reward": 0.
|
| 34 |
-
"reward_std": 0.
|
| 35 |
-
"rewards/accuracy_reward": 0.
|
| 36 |
-
"rewards/format_reward": 0.
|
| 37 |
"step": 5
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"clip_ratio": 0.0,
|
| 41 |
-
"completion_length":
|
| 42 |
"epoch": 0.17057569296375266,
|
| 43 |
-
"grad_norm": 0.
|
| 44 |
-
"kl": 0.
|
| 45 |
"learning_rate": 1.25e-06,
|
| 46 |
-
"loss": 0.
|
| 47 |
-
"reward": 0.
|
| 48 |
-
"reward_std": 0.
|
| 49 |
-
"rewards/accuracy_reward": 0.
|
| 50 |
-
"rewards/format_reward": 0.
|
| 51 |
"step": 10
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"clip_ratio": 0.0,
|
| 55 |
-
"completion_length":
|
| 56 |
"epoch": 0.255863539445629,
|
| 57 |
-
"grad_norm":
|
| 58 |
-
"kl": 0.
|
| 59 |
"learning_rate": 1.875e-06,
|
| 60 |
-
"loss": 0.
|
| 61 |
-
"reward": 0.
|
| 62 |
-
"reward_std": 0.
|
| 63 |
-
"rewards/accuracy_reward": 0.
|
| 64 |
-
"rewards/format_reward": 0.
|
| 65 |
"step": 15
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"clip_ratio": 0.0,
|
| 69 |
-
"completion_length":
|
| 70 |
"epoch": 0.3411513859275053,
|
| 71 |
-
"grad_norm": 0.
|
| 72 |
-
"kl": 0.
|
| 73 |
"learning_rate": 2.5e-06,
|
| 74 |
-
"loss": 0.
|
| 75 |
-
"reward": 1.
|
| 76 |
-
"reward_std": 0.
|
| 77 |
-
"rewards/accuracy_reward": 0.
|
| 78 |
-
"rewards/format_reward": 0.
|
| 79 |
"step": 20
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"clip_ratio": 0.0,
|
| 83 |
-
"completion_length":
|
| 84 |
"epoch": 0.42643923240938164,
|
| 85 |
-
"grad_norm":
|
| 86 |
-
"kl": 0.
|
| 87 |
"learning_rate": 2.999828909426247e-06,
|
| 88 |
-
"loss": 0.
|
| 89 |
-
"reward": 1.
|
| 90 |
-
"reward_std": 0.
|
| 91 |
-
"rewards/accuracy_reward": 0.
|
| 92 |
-
"rewards/format_reward": 0.
|
| 93 |
"step": 25
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"clip_ratio": 0.0,
|
| 97 |
-
"completion_length":
|
| 98 |
"epoch": 0.511727078891258,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"kl": 0.
|
| 101 |
"learning_rate": 2.9938448364256362e-06,
|
| 102 |
-
"loss": 0.
|
| 103 |
-
"reward": 1.
|
| 104 |
-
"reward_std": 0.
|
| 105 |
-
"rewards/accuracy_reward": 0.
|
| 106 |
-
"rewards/format_reward": 0.
|
| 107 |
"step": 30
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"clip_ratio": 0.0,
|
| 111 |
-
"completion_length":
|
| 112 |
"epoch": 0.5970149253731343,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"kl":
|
| 115 |
"learning_rate": 2.979345224048116e-06,
|
| 116 |
-
"loss":
|
| 117 |
-
"reward": 1.
|
| 118 |
-
"reward_std": 0.
|
| 119 |
-
"rewards/accuracy_reward": 0.
|
| 120 |
-
"rewards/format_reward": 0.
|
| 121 |
"step": 35
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"clip_ratio": 0.0,
|
| 125 |
-
"completion_length":
|
| 126 |
"epoch": 0.6823027718550106,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"kl":
|
| 129 |
"learning_rate": 2.956412726139078e-06,
|
| 130 |
-
"loss":
|
| 131 |
-
"reward": 1.
|
| 132 |
-
"reward_std": 0.
|
| 133 |
-
"rewards/accuracy_reward": 0.
|
| 134 |
-
"rewards/format_reward": 0.
|
| 135 |
"step": 40
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"clip_ratio": 0.0,
|
| 139 |
-
"completion_length":
|
| 140 |
"epoch": 0.767590618336887,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"kl":
|
| 143 |
"learning_rate": 2.925178067512904e-06,
|
| 144 |
-
"loss":
|
| 145 |
-
"reward": 1.
|
| 146 |
-
"reward_std": 0.
|
| 147 |
-
"rewards/accuracy_reward": 0.
|
| 148 |
-
"rewards/format_reward": 0.
|
| 149 |
"step": 45
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"clip_ratio": 0.0,
|
| 153 |
-
"completion_length":
|
| 154 |
"epoch": 0.8528784648187633,
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"kl":
|
| 157 |
"learning_rate": 2.88581929876693e-06,
|
| 158 |
-
"loss": 0.
|
| 159 |
-
"reward":
|
| 160 |
-
"reward_std": 0.
|
| 161 |
-
"rewards/accuracy_reward": 0.
|
| 162 |
-
"rewards/format_reward": 0.
|
| 163 |
"step": 50
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"clip_ratio": 0.0,
|
| 167 |
-
"completion_length":
|
| 168 |
"epoch": 0.9381663113006397,
|
| 169 |
-
"grad_norm":
|
| 170 |
-
"kl":
|
| 171 |
"learning_rate": 2.8385607813186967e-06,
|
| 172 |
-
"loss": 0.
|
| 173 |
-
"reward":
|
| 174 |
-
"reward_std": 0.
|
| 175 |
-
"rewards/accuracy_reward": 0.
|
| 176 |
-
"rewards/format_reward": 0.
|
| 177 |
"step": 55
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"clip_ratio": 0.0,
|
| 181 |
-
"completion_length":
|
| 182 |
"epoch": 1.0341151385927505,
|
| 183 |
-
"grad_norm":
|
| 184 |
-
"kl":
|
| 185 |
"learning_rate": 2.7836719084521715e-06,
|
| 186 |
-
"loss": 0.
|
| 187 |
-
"reward": 1.
|
| 188 |
-
"reward_std": 0.
|
| 189 |
-
"rewards/accuracy_reward": 0.
|
| 190 |
-
"rewards/format_reward": 0.
|
| 191 |
"step": 60
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"clip_ratio": 0.0,
|
| 195 |
-
"completion_length":
|
| 196 |
"epoch": 1.1194029850746268,
|
| 197 |
-
"grad_norm":
|
| 198 |
-
"kl":
|
| 199 |
"learning_rate": 2.7214655696635407e-06,
|
| 200 |
-
"loss": 0.
|
| 201 |
-
"reward": 1.
|
| 202 |
-
"reward_std": 0.
|
| 203 |
-
"rewards/accuracy_reward": 0.
|
| 204 |
-
"rewards/format_reward": 0.
|
| 205 |
"step": 65
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"clip_ratio": 0.0,
|
| 209 |
-
"completion_length":
|
| 210 |
"epoch": 1.2046908315565032,
|
| 211 |
-
"grad_norm":
|
| 212 |
-
"kl": 2.
|
| 213 |
"learning_rate": 2.652296367060421e-06,
|
| 214 |
-
"loss": 0.
|
| 215 |
-
"reward": 1.
|
| 216 |
-
"reward_std": 0.
|
| 217 |
-
"rewards/accuracy_reward": 0.
|
| 218 |
-
"rewards/format_reward": 0.
|
| 219 |
"step": 70
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"clip_ratio": 0.0,
|
| 223 |
-
"completion_length":
|
| 224 |
"epoch": 1.2899786780383795,
|
| 225 |
-
"grad_norm":
|
| 226 |
-
"kl":
|
| 227 |
"learning_rate": 2.5765585939817676e-06,
|
| 228 |
-
"loss": 0.
|
| 229 |
-
"reward": 1.
|
| 230 |
-
"reward_std": 0.
|
| 231 |
-
"rewards/accuracy_reward": 0.
|
| 232 |
-
"rewards/format_reward": 0.
|
| 233 |
"step": 75
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"clip_ratio": 0.0,
|
| 237 |
-
"completion_length":
|
| 238 |
"epoch": 1.375266524520256,
|
| 239 |
-
"grad_norm":
|
| 240 |
-
"kl":
|
| 241 |
"learning_rate": 2.4946839873611927e-06,
|
| 242 |
-
"loss": 0.
|
| 243 |
-
"reward": 1.
|
| 244 |
-
"reward_std": 0.
|
| 245 |
-
"rewards/accuracy_reward": 0.
|
| 246 |
-
"rewards/format_reward": 0.
|
| 247 |
"step": 80
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"clip_ratio": 0.0,
|
| 251 |
-
"completion_length":
|
| 252 |
"epoch": 1.4605543710021323,
|
| 253 |
-
"grad_norm":
|
| 254 |
-
"kl":
|
| 255 |
"learning_rate": 2.4071392666461563e-06,
|
| 256 |
-
"loss": 0.
|
| 257 |
-
"reward": 1.
|
| 258 |
-
"reward_std": 0.
|
| 259 |
-
"rewards/accuracy_reward": 0.
|
| 260 |
-
"rewards/format_reward": 0.
|
| 261 |
"step": 85
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"clip_ratio": 0.0,
|
| 265 |
-
"completion_length":
|
| 266 |
"epoch": 1.5458422174840085,
|
| 267 |
-
"grad_norm":
|
| 268 |
-
"kl":
|
| 269 |
"learning_rate": 2.314423473302218e-06,
|
| 270 |
-
"loss": 0.
|
| 271 |
-
"reward": 1.
|
| 272 |
-
"reward_std": 0.
|
| 273 |
-
"rewards/accuracy_reward": 0.
|
| 274 |
-
"rewards/format_reward": 0.
|
| 275 |
"step": 90
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"clip_ratio": 0.0,
|
| 279 |
-
"completion_length":
|
| 280 |
"epoch": 1.6311300639658848,
|
| 281 |
-
"grad_norm":
|
| 282 |
-
"kl":
|
| 283 |
"learning_rate": 2.2170651260682927e-06,
|
| 284 |
-
"loss": 0.
|
| 285 |
-
"reward": 1.
|
| 286 |
-
"reward_std": 0.
|
| 287 |
-
"rewards/accuracy_reward": 0.
|
| 288 |
-
"rewards/format_reward": 0.
|
| 289 |
"step": 95
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 1.716417910447761,
|
| 293 |
-
"grad_norm":
|
| 294 |
"learning_rate": 2.1156192081791355e-06,
|
| 295 |
-
"loss":
|
| 296 |
"step": 100
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 1.716417910447761,
|
| 300 |
"eval_clip_ratio": 0.0,
|
| 301 |
-
"eval_completion_length":
|
| 302 |
-
"eval_kl":
|
| 303 |
-
"eval_loss": 0.
|
| 304 |
-
"eval_reward": 1.
|
| 305 |
-
"eval_reward_std": 0.
|
| 306 |
-
"eval_rewards/accuracy_reward": 0.
|
| 307 |
-
"eval_rewards/format_reward": 0.
|
| 308 |
-
"eval_runtime":
|
| 309 |
-
"eval_samples_per_second": 1.
|
| 310 |
"eval_steps_per_second": 0.01,
|
| 311 |
"step": 100
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"clip_ratio": 0.0,
|
| 315 |
-
"completion_length":
|
| 316 |
"epoch": 1.8017057569296375,
|
| 317 |
-
"grad_norm":
|
| 318 |
-
"kl":
|
| 319 |
"learning_rate": 2.010664003729149e-06,
|
| 320 |
-
"loss": 0.
|
| 321 |
-
"reward": 1.
|
| 322 |
-
"reward_std": 0.
|
| 323 |
-
"rewards/accuracy_reward": 0.
|
| 324 |
-
"rewards/format_reward": 0.
|
| 325 |
"step": 105
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"clip_ratio": 0.0,
|
| 329 |
-
"completion_length":
|
| 330 |
"epoch": 1.886993603411514,
|
| 331 |
-
"grad_norm":
|
| 332 |
-
"kl":
|
| 333 |
"learning_rate": 1.9027978012115653e-06,
|
| 334 |
-
"loss": 0.
|
| 335 |
-
"reward": 1.
|
| 336 |
-
"reward_std": 0.
|
| 337 |
-
"rewards/accuracy_reward": 0.
|
| 338 |
-
"rewards/format_reward": 0.
|
| 339 |
"step": 110
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"clip_ratio": 0.0,
|
| 343 |
-
"completion_length":
|
| 344 |
"epoch": 1.9722814498933903,
|
| 345 |
-
"grad_norm":
|
| 346 |
-
"kl":
|
| 347 |
"learning_rate": 1.7926354830241926e-06,
|
| 348 |
-
"loss": 0.
|
| 349 |
-
"reward": 1.
|
| 350 |
-
"reward_std": 0.
|
| 351 |
-
"rewards/accuracy_reward": 0.
|
| 352 |
-
"rewards/format_reward": 0.
|
| 353 |
"step": 115
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"clip_ratio": 0.0,
|
| 357 |
-
"completion_length":
|
| 358 |
"epoch": 2.068230277185501,
|
| 359 |
-
"grad_norm":
|
| 360 |
-
"kl":
|
| 361 |
"learning_rate": 1.6808050203829845e-06,
|
| 362 |
-
"loss": 0.
|
| 363 |
-
"reward": 1.
|
| 364 |
-
"reward_std": 0.
|
| 365 |
-
"rewards/accuracy_reward": 0.
|
| 366 |
-
"rewards/format_reward": 0.
|
| 367 |
"step": 120
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"clip_ratio": 0.0,
|
| 371 |
-
"completion_length":
|
| 372 |
"epoch": 2.1535181236673773,
|
| 373 |
-
"grad_norm":
|
| 374 |
-
"kl": 3.
|
| 375 |
"learning_rate": 1.5679438936238768e-06,
|
| 376 |
-
"loss": 0.
|
| 377 |
-
"reward": 1.
|
| 378 |
-
"reward_std": 0.
|
| 379 |
-
"rewards/accuracy_reward": 0.
|
| 380 |
-
"rewards/format_reward": 0.
|
| 381 |
"step": 125
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"clip_ratio": 0.0,
|
| 385 |
-
"completion_length":
|
| 386 |
"epoch": 2.2388059701492535,
|
| 387 |
-
"grad_norm":
|
| 388 |
-
"kl":
|
| 389 |
"learning_rate": 1.454695458298667e-06,
|
| 390 |
-
"loss": 0.
|
| 391 |
-
"reward": 1.
|
| 392 |
-
"reward_std": 0.
|
| 393 |
-
"rewards/accuracy_reward": 0.
|
| 394 |
-
"rewards/format_reward": 0.
|
| 395 |
"step": 130
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"clip_ratio": 0.0,
|
| 399 |
-
"completion_length":
|
| 400 |
"epoch": 2.3240938166311302,
|
| 401 |
-
"grad_norm":
|
| 402 |
-
"kl":
|
| 403 |
"learning_rate": 1.341705277779715e-06,
|
| 404 |
-
"loss": 0.
|
| 405 |
-
"reward": 1.
|
| 406 |
-
"reward_std": 0.
|
| 407 |
-
"rewards/accuracy_reward": 0.
|
| 408 |
-
"rewards/format_reward": 0.
|
| 409 |
"step": 135
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"clip_ratio": 0.0,
|
| 413 |
-
"completion_length":
|
| 414 |
"epoch": 2.4093816631130065,
|
| 415 |
-
"grad_norm":
|
| 416 |
-
"kl":
|
| 417 |
"learning_rate": 1.2296174432791415e-06,
|
| 418 |
-
"loss": 0.
|
| 419 |
-
"reward": 1.
|
| 420 |
-
"reward_std": 0.
|
| 421 |
-
"rewards/accuracy_reward": 0.
|
| 422 |
-
"rewards/format_reward": 0.
|
| 423 |
"step": 140
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"clip_ratio": 0.0,
|
| 427 |
-
"completion_length":
|
| 428 |
"epoch": 2.4946695095948828,
|
| 429 |
-
"grad_norm":
|
| 430 |
-
"kl":
|
| 431 |
"learning_rate": 1.1190709022599545e-06,
|
| 432 |
-
"loss": 0.
|
| 433 |
-
"reward": 1.
|
| 434 |
-
"reward_std": 0.
|
| 435 |
-
"rewards/accuracy_reward": 0.
|
| 436 |
-
"rewards/format_reward": 0.
|
| 437 |
"step": 145
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"clip_ratio": 0.0,
|
| 441 |
-
"completion_length":
|
| 442 |
"epoch": 2.579957356076759,
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"kl":
|
| 445 |
"learning_rate": 1.0106958161686963e-06,
|
| 446 |
-
"loss": 0.
|
| 447 |
-
"reward": 1.
|
| 448 |
-
"reward_std": 0.
|
| 449 |
-
"rewards/accuracy_reward": 0.
|
| 450 |
-
"rewards/format_reward": 0.
|
| 451 |
"step": 150
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"clip_ratio": 0.0,
|
| 455 |
-
"completion_length":
|
| 456 |
"epoch": 2.6652452025586353,
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"kl":
|
| 459 |
"learning_rate": 9.051099682520474e-07,
|
| 460 |
-
"loss": 0.
|
| 461 |
-
"reward": 1.
|
| 462 |
-
"reward_std": 0.
|
| 463 |
-
"rewards/accuracy_reward": 0.
|
| 464 |
-
"rewards/format_reward": 0.
|
| 465 |
"step": 155
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"clip_ratio": 0.0,
|
| 469 |
-
"completion_length":
|
| 470 |
"epoch": 2.750533049040512,
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"kl": 2.
|
| 473 |
"learning_rate": 8.029152419343472e-07,
|
| 474 |
-
"loss": 0.
|
| 475 |
-
"reward": 1.
|
| 476 |
-
"reward_std": 0.
|
| 477 |
-
"rewards/accuracy_reward": 0.
|
| 478 |
-
"rewards/format_reward": 0.
|
| 479 |
"step": 160
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"clip_ratio": 0.0,
|
| 483 |
-
"completion_length":
|
| 484 |
"epoch": 2.835820895522388,
|
| 485 |
-
"grad_norm":
|
| 486 |
-
"kl": 2.
|
| 487 |
"learning_rate": 7.046941898307347e-07,
|
| 488 |
-
"loss": 0.
|
| 489 |
-
"reward": 1.
|
| 490 |
-
"reward_std": 0.
|
| 491 |
-
"rewards/accuracy_reward": 0.
|
| 492 |
-
"rewards/format_reward": 0.
|
| 493 |
"step": 165
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"clip_ratio": 0.0,
|
| 497 |
-
"completion_length":
|
| 498 |
"epoch": 2.9211087420042645,
|
| 499 |
-
"grad_norm":
|
| 500 |
-
"kl":
|
| 501 |
"learning_rate": 6.11006712953975e-07,
|
| 502 |
-
"loss": 0.
|
| 503 |
-
"reward": 1.
|
| 504 |
-
"reward_std": 0.
|
| 505 |
-
"rewards/accuracy_reward": 0.
|
| 506 |
-
"rewards/format_reward": 0.
|
| 507 |
"step": 170
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"clip_ratio": 0.0,
|
| 511 |
-
"completion_length":
|
| 512 |
"epoch": 3.0170575692963753,
|
| 513 |
-
"grad_norm":
|
| 514 |
-
"kl": 3.
|
| 515 |
"learning_rate": 5.223868690448817e-07,
|
| 516 |
-
"loss": 0.
|
| 517 |
-
"reward": 1.
|
| 518 |
-
"reward_std": 0.
|
| 519 |
-
"rewards/accuracy_reward": 0.
|
| 520 |
-
"rewards/format_reward": 0.
|
| 521 |
"step": 175
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"clip_ratio": 0.0,
|
| 525 |
-
"completion_length":
|
| 526 |
"epoch": 3.1023454157782515,
|
| 527 |
-
"grad_norm":
|
| 528 |
-
"kl": 3.
|
| 529 |
"learning_rate": 4.3933982822017883e-07,
|
| 530 |
-
"loss": 0.
|
| 531 |
-
"reward": 1.
|
| 532 |
-
"reward_std": 0.
|
| 533 |
-
"rewards/accuracy_reward": 0.
|
| 534 |
-
"rewards/format_reward": 0.
|
| 535 |
"step": 180
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"clip_ratio": 0.0,
|
| 539 |
-
"completion_length":
|
| 540 |
"epoch": 3.1876332622601278,
|
| 541 |
-
"grad_norm":
|
| 542 |
-
"kl":
|
| 543 |
"learning_rate": 3.6233899329188115e-07,
|
| 544 |
-
"loss":
|
| 545 |
-
"reward": 1.
|
| 546 |
-
"reward_std": 0.
|
| 547 |
-
"rewards/accuracy_reward": 0.
|
| 548 |
-
"rewards/format_reward": 0.
|
| 549 |
"step": 185
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"clip_ratio": 0.0,
|
| 553 |
-
"completion_length":
|
| 554 |
"epoch": 3.272921108742004,
|
| 555 |
-
"grad_norm":
|
| 556 |
-
"kl":
|
| 557 |
"learning_rate": 2.9182330117358096e-07,
|
| 558 |
-
"loss":
|
| 559 |
-
"reward": 1.
|
| 560 |
-
"reward_std": 0.
|
| 561 |
-
"rewards/accuracy_reward": 0.
|
| 562 |
-
"rewards/format_reward": 0.
|
| 563 |
"step": 190
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"clip_ratio": 0.0,
|
| 567 |
-
"completion_length":
|
| 568 |
"epoch": 3.3582089552238807,
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"kl":
|
| 571 |
"learning_rate": 2.281947207567473e-07,
|
| 572 |
-
"loss": 0.
|
| 573 |
-
"reward": 1.
|
| 574 |
-
"reward_std": 0.
|
| 575 |
-
"rewards/accuracy_reward": 0.
|
| 576 |
-
"rewards/format_reward": 0.
|
| 577 |
"step": 195
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 3.443496801705757,
|
| 581 |
-
"grad_norm":
|
| 582 |
"learning_rate": 1.718159615201853e-07,
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 200
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 3.443496801705757,
|
| 588 |
"eval_clip_ratio": 0.0,
|
| 589 |
-
"eval_completion_length":
|
| 590 |
-
"eval_kl": 2.
|
| 591 |
-
"eval_loss": 0.
|
| 592 |
-
"eval_reward": 1.
|
| 593 |
-
"eval_reward_std": 0.
|
| 594 |
-
"eval_rewards/accuracy_reward": 0.
|
| 595 |
-
"eval_rewards/format_reward": 0.
|
| 596 |
-
"eval_runtime":
|
| 597 |
-
"eval_samples_per_second": 1.
|
| 598 |
"eval_steps_per_second": 0.011,
|
| 599 |
"step": 200
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"clip_ratio": 0.0,
|
| 603 |
-
"completion_length":
|
| 604 |
"epoch": 3.5287846481876333,
|
| 605 |
-
"grad_norm":
|
| 606 |
-
"kl": 2.
|
| 607 |
"learning_rate": 1.2300840593454622e-07,
|
| 608 |
-
"loss": 0.
|
| 609 |
-
"reward": 1.
|
| 610 |
-
"reward_std": 0.
|
| 611 |
-
"rewards/accuracy_reward": 0.
|
| 612 |
-
"rewards/format_reward": 0.
|
| 613 |
"step": 205
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"clip_ratio": 0.0,
|
| 617 |
-
"completion_length":
|
| 618 |
"epoch": 3.6140724946695095,
|
| 619 |
-
"grad_norm":
|
| 620 |
-
"kl": 2.
|
| 621 |
"learning_rate": 8.20502774480395e-08,
|
| 622 |
-
"loss": 0.
|
| 623 |
-
"reward": 1.
|
| 624 |
-
"reward_std": 0.
|
| 625 |
-
"rewards/accuracy_reward": 0.
|
| 626 |
-
"rewards/format_reward": 0.
|
| 627 |
"step": 210
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"clip_ratio": 0.0,
|
| 631 |
-
"completion_length":
|
| 632 |
"epoch": 3.699360341151386,
|
| 633 |
-
"grad_norm":
|
| 634 |
-
"kl": 2.
|
| 635 |
"learning_rate": 4.917505449659615e-08,
|
| 636 |
-
"loss": 0.
|
| 637 |
-
"reward": 1.
|
| 638 |
-
"reward_std": 0.
|
| 639 |
-
"rewards/accuracy_reward": 0.
|
| 640 |
-
"rewards/format_reward": 0.
|
| 641 |
"step": 215
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"clip_ratio": 0.0,
|
| 645 |
-
"completion_length":
|
| 646 |
"epoch": 3.7846481876332625,
|
| 647 |
-
"grad_norm":
|
| 648 |
-
"kl": 2.
|
| 649 |
"learning_rate": 2.4570139579284723e-08,
|
| 650 |
-
"loss": 0.
|
| 651 |
-
"reward": 1.
|
| 652 |
-
"reward_std": 0.
|
| 653 |
-
"rewards/accuracy_reward": 0.
|
| 654 |
-
"rewards/format_reward": 0.
|
| 655 |
"step": 220
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"clip_ratio": 0.0,
|
| 659 |
-
"completion_length":
|
| 660 |
"epoch": 3.8699360341151388,
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"kl": 2.
|
| 663 |
"learning_rate": 8.37579098581176e-09,
|
| 664 |
-
"loss": 0.
|
| 665 |
-
"reward": 1.
|
| 666 |
-
"reward_std": 0.
|
| 667 |
-
"rewards/accuracy_reward": 0.
|
| 668 |
-
"rewards/format_reward": 0.
|
| 669 |
"step": 225
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"clip_ratio": 0.0,
|
| 673 |
-
"completion_length":
|
| 674 |
"epoch": 3.955223880597015,
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"kl": 2.
|
| 677 |
"learning_rate": 6.843232656998933e-10,
|
| 678 |
-
"loss": 0.
|
| 679 |
-
"reward": 1.
|
| 680 |
-
"reward_std": 0.
|
| 681 |
-
"rewards/accuracy_reward": 0.
|
| 682 |
-
"rewards/format_reward": 0.
|
| 683 |
"step": 230
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"clip_ratio": 0.0,
|
| 687 |
-
"completion_length":
|
| 688 |
"epoch": 3.9893390191897655,
|
| 689 |
-
"kl": 2.
|
| 690 |
-
"reward": 1.
|
| 691 |
-
"reward_std": 0.
|
| 692 |
-
"rewards/accuracy_reward": 0.
|
| 693 |
-
"rewards/format_reward": 0.
|
| 694 |
"step": 232,
|
| 695 |
"total_flos": 0.0,
|
| 696 |
-
"train_loss":
|
| 697 |
-
"train_runtime":
|
| 698 |
-
"train_samples_per_second": 0.
|
| 699 |
"train_steps_per_second": 0.006
|
| 700 |
}
|
| 701 |
],
|
|
|
|
| 12 |
"clip_ratio": 0.0,
|
| 13 |
"completion_length": 445.9810485839844,
|
| 14 |
"epoch": 0.017057569296375266,
|
| 15 |
+
"grad_norm": 0.4470120966434479,
|
| 16 |
"kl": 0.0,
|
| 17 |
"learning_rate": 1.25e-07,
|
| 18 |
"loss": -0.0061,
|
|
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"clip_ratio": 0.0,
|
| 27 |
+
"completion_length": 457.7648096084595,
|
| 28 |
"epoch": 0.08528784648187633,
|
| 29 |
+
"grad_norm": 0.46972450613975525,
|
| 30 |
+
"kl": 4.832446575164795e-05,
|
| 31 |
"learning_rate": 6.25e-07,
|
| 32 |
+
"loss": -0.0007,
|
| 33 |
+
"reward": 0.2547433148138225,
|
| 34 |
+
"reward_std": 0.3450065036304295,
|
| 35 |
+
"rewards/accuracy_reward": 0.19698661542497575,
|
| 36 |
+
"rewards/format_reward": 0.05775669927243143,
|
| 37 |
"step": 5
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"clip_ratio": 0.0,
|
| 41 |
+
"completion_length": 449.00381622314455,
|
| 42 |
"epoch": 0.17057569296375266,
|
| 43 |
+
"grad_norm": 0.4081119894981384,
|
| 44 |
+
"kl": 0.0028772354125976562,
|
| 45 |
"learning_rate": 1.25e-06,
|
| 46 |
+
"loss": 0.0138,
|
| 47 |
+
"reward": 0.3084821570664644,
|
| 48 |
+
"reward_std": 0.38891434073448183,
|
| 49 |
+
"rewards/accuracy_reward": 0.2015625100582838,
|
| 50 |
+
"rewards/format_reward": 0.10691964821889996,
|
| 51 |
"step": 10
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"clip_ratio": 0.0,
|
| 55 |
+
"completion_length": 445.6658676147461,
|
| 56 |
"epoch": 0.255863539445629,
|
| 57 |
+
"grad_norm": 0.8681241273880005,
|
| 58 |
+
"kl": 0.05352325439453125,
|
| 59 |
"learning_rate": 1.875e-06,
|
| 60 |
+
"loss": 0.0257,
|
| 61 |
+
"reward": 0.6531250305473805,
|
| 62 |
+
"reward_std": 0.5433857575058937,
|
| 63 |
+
"rewards/accuracy_reward": 0.25892858393490314,
|
| 64 |
+
"rewards/format_reward": 0.3941964443773031,
|
| 65 |
"step": 15
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"clip_ratio": 0.0,
|
| 69 |
+
"completion_length": 469.591983795166,
|
| 70 |
"epoch": 0.3411513859275053,
|
| 71 |
+
"grad_norm": 0.5079839825630188,
|
| 72 |
+
"kl": 0.0329986572265625,
|
| 73 |
"learning_rate": 2.5e-06,
|
| 74 |
+
"loss": 0.0307,
|
| 75 |
+
"reward": 1.051785758137703,
|
| 76 |
+
"reward_std": 0.5734383672475815,
|
| 77 |
+
"rewards/accuracy_reward": 0.4296875223517418,
|
| 78 |
+
"rewards/format_reward": 0.6220982387661934,
|
| 79 |
"step": 20
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"clip_ratio": 0.0,
|
| 83 |
+
"completion_length": 494.53462371826174,
|
| 84 |
"epoch": 0.42643923240938164,
|
| 85 |
+
"grad_norm": 0.7165977954864502,
|
| 86 |
+
"kl": 0.0740936279296875,
|
| 87 |
"learning_rate": 2.999828909426247e-06,
|
| 88 |
+
"loss": 0.046,
|
| 89 |
+
"reward": 1.3908482611179351,
|
| 90 |
+
"reward_std": 0.45604070723056794,
|
| 91 |
+
"rewards/accuracy_reward": 0.5560268141329289,
|
| 92 |
+
"rewards/format_reward": 0.8348214745521545,
|
| 93 |
"step": 25
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"clip_ratio": 0.0,
|
| 97 |
+
"completion_length": 466.7975654602051,
|
| 98 |
"epoch": 0.511727078891258,
|
| 99 |
+
"grad_norm": 105.83580780029297,
|
| 100 |
+
"kl": 0.15655517578125,
|
| 101 |
"learning_rate": 2.9938448364256362e-06,
|
| 102 |
+
"loss": 0.0562,
|
| 103 |
+
"reward": 1.4703125596046447,
|
| 104 |
+
"reward_std": 0.39982456117868426,
|
| 105 |
+
"rewards/accuracy_reward": 0.5502232410013675,
|
| 106 |
+
"rewards/format_reward": 0.9200893312692642,
|
| 107 |
"step": 30
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"clip_ratio": 0.0,
|
| 111 |
+
"completion_length": 506.7870819091797,
|
| 112 |
"epoch": 0.5970149253731343,
|
| 113 |
+
"grad_norm": 185.41656494140625,
|
| 114 |
+
"kl": 0.2352294921875,
|
| 115 |
"learning_rate": 2.979345224048116e-06,
|
| 116 |
+
"loss": 0.1242,
|
| 117 |
+
"reward": 1.4352679133415223,
|
| 118 |
+
"reward_std": 0.4489331416785717,
|
| 119 |
+
"rewards/accuracy_reward": 0.5587053798139096,
|
| 120 |
+
"rewards/format_reward": 0.8765625402331352,
|
| 121 |
"step": 35
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"clip_ratio": 0.0,
|
| 125 |
+
"completion_length": 534.5277046203613,
|
| 126 |
"epoch": 0.6823027718550106,
|
| 127 |
+
"grad_norm": 174158.28125,
|
| 128 |
+
"kl": 1967.46162109375,
|
| 129 |
"learning_rate": 2.956412726139078e-06,
|
| 130 |
+
"loss": 65.1453,
|
| 131 |
+
"reward": 1.3604911327362061,
|
| 132 |
+
"reward_std": 0.5219059348106384,
|
| 133 |
+
"rewards/accuracy_reward": 0.557812524586916,
|
| 134 |
+
"rewards/format_reward": 0.8026786133646965,
|
| 135 |
"step": 40
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"clip_ratio": 0.0,
|
| 139 |
+
"completion_length": 597.9991287231445,
|
| 140 |
"epoch": 0.767590618336887,
|
| 141 |
+
"grad_norm": 19395.80078125,
|
| 142 |
+
"kl": 781.421875,
|
| 143 |
"learning_rate": 2.925178067512904e-06,
|
| 144 |
+
"loss": 24.4654,
|
| 145 |
+
"reward": 1.210491119325161,
|
| 146 |
+
"reward_std": 0.614486200362444,
|
| 147 |
+
"rewards/accuracy_reward": 0.5167410962283612,
|
| 148 |
+
"rewards/format_reward": 0.6937500357627868,
|
| 149 |
"step": 45
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"clip_ratio": 0.0,
|
| 153 |
+
"completion_length": 663.8980178833008,
|
| 154 |
"epoch": 0.8528784648187633,
|
| 155 |
+
"grad_norm": 12.801368713378906,
|
| 156 |
+
"kl": 8.7136962890625,
|
| 157 |
"learning_rate": 2.88581929876693e-06,
|
| 158 |
+
"loss": 0.4607,
|
| 159 |
+
"reward": 0.9968750447034835,
|
| 160 |
+
"reward_std": 0.6696360170841217,
|
| 161 |
+
"rewards/accuracy_reward": 0.4488839447498322,
|
| 162 |
+
"rewards/format_reward": 0.547991094738245,
|
| 163 |
"step": 50
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"clip_ratio": 0.0,
|
| 167 |
+
"completion_length": 697.5154357910156,
|
| 168 |
"epoch": 0.9381663113006397,
|
| 169 |
+
"grad_norm": 38.23894119262695,
|
| 170 |
+
"kl": 0.28189697265625,
|
| 171 |
"learning_rate": 2.8385607813186967e-06,
|
| 172 |
+
"loss": 0.2599,
|
| 173 |
+
"reward": 0.9542411118745804,
|
| 174 |
+
"reward_std": 0.6912495926022529,
|
| 175 |
+
"rewards/accuracy_reward": 0.4694196626543999,
|
| 176 |
+
"rewards/format_reward": 0.4848214492201805,
|
| 177 |
"step": 55
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"clip_ratio": 0.0,
|
| 181 |
+
"completion_length": 649.4566436767578,
|
| 182 |
"epoch": 1.0341151385927505,
|
| 183 |
+
"grad_norm": 313.91156005859375,
|
| 184 |
+
"kl": 4.05673828125,
|
| 185 |
"learning_rate": 2.7836719084521715e-06,
|
| 186 |
+
"loss": 0.5276,
|
| 187 |
+
"reward": 1.1421875596046447,
|
| 188 |
+
"reward_std": 0.7461830854415894,
|
| 189 |
+
"rewards/accuracy_reward": 0.529910734295845,
|
| 190 |
+
"rewards/format_reward": 0.6122768133878708,
|
| 191 |
"step": 60
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"clip_ratio": 0.0,
|
| 195 |
+
"completion_length": 657.656948852539,
|
| 196 |
"epoch": 1.1194029850746268,
|
| 197 |
+
"grad_norm": 83.17515563964844,
|
| 198 |
+
"kl": 4.4130859375,
|
| 199 |
"learning_rate": 2.7214655696635407e-06,
|
| 200 |
+
"loss": 0.5723,
|
| 201 |
+
"reward": 1.0674107655882836,
|
| 202 |
+
"reward_std": 0.7840202301740646,
|
| 203 |
+
"rewards/accuracy_reward": 0.4863839507102966,
|
| 204 |
+
"rewards/format_reward": 0.5810268193483352,
|
| 205 |
"step": 65
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"clip_ratio": 0.0,
|
| 209 |
+
"completion_length": 586.2384178161622,
|
| 210 |
"epoch": 1.2046908315565032,
|
| 211 |
+
"grad_norm": 15.149043083190918,
|
| 212 |
+
"kl": 2.45390625,
|
| 213 |
"learning_rate": 2.652296367060421e-06,
|
| 214 |
+
"loss": 0.4104,
|
| 215 |
+
"reward": 1.256919699907303,
|
| 216 |
+
"reward_std": 0.7236043408513069,
|
| 217 |
+
"rewards/accuracy_reward": 0.5531250238418579,
|
| 218 |
+
"rewards/format_reward": 0.703794677555561,
|
| 219 |
"step": 70
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"clip_ratio": 0.0,
|
| 223 |
+
"completion_length": 574.1111854553222,
|
| 224 |
"epoch": 1.2899786780383795,
|
| 225 |
+
"grad_norm": 21.18400764465332,
|
| 226 |
+
"kl": 9.9931640625,
|
| 227 |
"learning_rate": 2.5765585939817676e-06,
|
| 228 |
+
"loss": 0.744,
|
| 229 |
+
"reward": 1.3468750655651092,
|
| 230 |
+
"reward_std": 0.6585553318262101,
|
| 231 |
+
"rewards/accuracy_reward": 0.5841518118977547,
|
| 232 |
+
"rewards/format_reward": 0.7627232521772385,
|
| 233 |
"step": 75
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"clip_ratio": 0.0,
|
| 237 |
+
"completion_length": 564.7518135070801,
|
| 238 |
"epoch": 1.375266524520256,
|
| 239 |
+
"grad_norm": 486.4627990722656,
|
| 240 |
+
"kl": 8.6583984375,
|
| 241 |
"learning_rate": 2.4946839873611927e-06,
|
| 242 |
+
"loss": 0.531,
|
| 243 |
+
"reward": 1.4149554252624512,
|
| 244 |
+
"reward_std": 0.6240757808089257,
|
| 245 |
+
"rewards/accuracy_reward": 0.6238839633762836,
|
| 246 |
+
"rewards/format_reward": 0.7910714626312256,
|
| 247 |
"step": 80
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"clip_ratio": 0.0,
|
| 251 |
+
"completion_length": 565.4381927490234,
|
| 252 |
"epoch": 1.4605543710021323,
|
| 253 |
+
"grad_norm": 88.8906021118164,
|
| 254 |
+
"kl": 2.0404296875,
|
| 255 |
"learning_rate": 2.4071392666461563e-06,
|
| 256 |
+
"loss": 0.1685,
|
| 257 |
+
"reward": 1.3975447028875352,
|
| 258 |
+
"reward_std": 0.6175315536558628,
|
| 259 |
+
"rewards/accuracy_reward": 0.6136160999536514,
|
| 260 |
+
"rewards/format_reward": 0.7839286029338837,
|
| 261 |
"step": 85
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"clip_ratio": 0.0,
|
| 265 |
+
"completion_length": 569.019669342041,
|
| 266 |
"epoch": 1.5458422174840085,
|
| 267 |
+
"grad_norm": 17.572298049926758,
|
| 268 |
+
"kl": 1.3623046875,
|
| 269 |
"learning_rate": 2.314423473302218e-06,
|
| 270 |
+
"loss": 0.0949,
|
| 271 |
+
"reward": 1.3495536297559738,
|
| 272 |
+
"reward_std": 0.6313827067613602,
|
| 273 |
+
"rewards/accuracy_reward": 0.5805803835391998,
|
| 274 |
+
"rewards/format_reward": 0.7689732491970063,
|
| 275 |
"step": 90
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"clip_ratio": 0.0,
|
| 279 |
+
"completion_length": 589.0908721923828,
|
| 280 |
"epoch": 1.6311300639658848,
|
| 281 |
+
"grad_norm": 219.19049072265625,
|
| 282 |
+
"kl": 3.03681640625,
|
| 283 |
"learning_rate": 2.2170651260682927e-06,
|
| 284 |
+
"loss": 0.178,
|
| 285 |
+
"reward": 1.3450893461704254,
|
| 286 |
+
"reward_std": 0.6524915762245656,
|
| 287 |
+
"rewards/accuracy_reward": 0.5834821671247482,
|
| 288 |
+
"rewards/format_reward": 0.7616071730852128,
|
| 289 |
"step": 95
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 1.716417910447761,
|
| 293 |
+
"grad_norm": 9490.1796875,
|
| 294 |
"learning_rate": 2.1156192081791355e-06,
|
| 295 |
+
"loss": 5.6995,
|
| 296 |
"step": 100
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 1.716417910447761,
|
| 300 |
"eval_clip_ratio": 0.0,
|
| 301 |
+
"eval_completion_length": 583.028025374245,
|
| 302 |
+
"eval_kl": 8.038014177316294,
|
| 303 |
+
"eval_loss": 0.30782344937324524,
|
| 304 |
+
"eval_reward": 1.30117531782522,
|
| 305 |
+
"eval_reward_std": 0.6690196339695599,
|
| 306 |
+
"eval_rewards/accuracy_reward": 0.538081950558641,
|
| 307 |
+
"eval_rewards/format_reward": 0.7630933717416879,
|
| 308 |
+
"eval_runtime": 4660.7236,
|
| 309 |
+
"eval_samples_per_second": 1.073,
|
| 310 |
"eval_steps_per_second": 0.01,
|
| 311 |
"step": 100
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"clip_ratio": 0.0,
|
| 315 |
+
"completion_length": 589.2927730560302,
|
| 316 |
"epoch": 1.8017057569296375,
|
| 317 |
+
"grad_norm": 36.098472595214844,
|
| 318 |
+
"kl": 46.948193359375,
|
| 319 |
"learning_rate": 2.010664003729149e-06,
|
| 320 |
+
"loss": 0.191,
|
| 321 |
+
"reward": 1.341964341700077,
|
| 322 |
+
"reward_std": 0.666185948625207,
|
| 323 |
+
"rewards/accuracy_reward": 0.579464315995574,
|
| 324 |
+
"rewards/format_reward": 0.7625000335276126,
|
| 325 |
"step": 105
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"clip_ratio": 0.0,
|
| 329 |
+
"completion_length": 572.2181030273438,
|
| 330 |
"epoch": 1.886993603411514,
|
| 331 |
+
"grad_norm": 33.85002517700195,
|
| 332 |
+
"kl": 1.8282470703125,
|
| 333 |
"learning_rate": 1.9027978012115653e-06,
|
| 334 |
+
"loss": 0.1948,
|
| 335 |
+
"reward": 1.3232143372297287,
|
| 336 |
+
"reward_std": 0.6623021572828293,
|
| 337 |
+
"rewards/accuracy_reward": 0.5607143074274064,
|
| 338 |
+
"rewards/format_reward": 0.7625000298023223,
|
| 339 |
"step": 110
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"clip_ratio": 0.0,
|
| 343 |
+
"completion_length": 542.2167655944825,
|
| 344 |
"epoch": 1.9722814498933903,
|
| 345 |
+
"grad_norm": 225.037109375,
|
| 346 |
+
"kl": 0.3978759765625,
|
| 347 |
"learning_rate": 1.7926354830241926e-06,
|
| 348 |
+
"loss": 0.0266,
|
| 349 |
+
"reward": 1.3252232730388642,
|
| 350 |
+
"reward_std": 0.6835980974137783,
|
| 351 |
+
"rewards/accuracy_reward": 0.5741071701049805,
|
| 352 |
+
"rewards/format_reward": 0.7511161074042321,
|
| 353 |
"step": 115
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"clip_ratio": 0.0,
|
| 357 |
+
"completion_length": 525.6891525268554,
|
| 358 |
"epoch": 2.068230277185501,
|
| 359 |
+
"grad_norm": 7.967476844787598,
|
| 360 |
+
"kl": 0.974072265625,
|
| 361 |
"learning_rate": 1.6808050203829845e-06,
|
| 362 |
+
"loss": 0.0252,
|
| 363 |
+
"reward": 1.304241132736206,
|
| 364 |
+
"reward_std": 0.6886323638260364,
|
| 365 |
+
"rewards/accuracy_reward": 0.5553571663796901,
|
| 366 |
+
"rewards/format_reward": 0.7488839626312256,
|
| 367 |
"step": 120
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"clip_ratio": 0.0,
|
| 371 |
+
"completion_length": 490.7185485839844,
|
| 372 |
"epoch": 2.1535181236673773,
|
| 373 |
+
"grad_norm": 2670.3564453125,
|
| 374 |
+
"kl": 3.83818359375,
|
| 375 |
"learning_rate": 1.5679438936238768e-06,
|
| 376 |
+
"loss": 0.0711,
|
| 377 |
+
"reward": 1.361607202887535,
|
| 378 |
+
"reward_std": 0.621609278768301,
|
| 379 |
+
"rewards/accuracy_reward": 0.5705357417464256,
|
| 380 |
+
"rewards/format_reward": 0.7910714641213417,
|
| 381 |
"step": 125
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"clip_ratio": 0.0,
|
| 385 |
+
"completion_length": 489.616316986084,
|
| 386 |
"epoch": 2.2388059701492535,
|
| 387 |
+
"grad_norm": 92.28889465332031,
|
| 388 |
+
"kl": 2.853662109375,
|
| 389 |
"learning_rate": 1.454695458298667e-06,
|
| 390 |
+
"loss": 0.0984,
|
| 391 |
+
"reward": 1.4082589894533157,
|
| 392 |
+
"reward_std": 0.5909771449863911,
|
| 393 |
+
"rewards/accuracy_reward": 0.5848214507102967,
|
| 394 |
+
"rewards/format_reward": 0.8234375342726707,
|
| 395 |
"step": 130
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"clip_ratio": 0.0,
|
| 399 |
+
"completion_length": 480.3973449707031,
|
| 400 |
"epoch": 2.3240938166311302,
|
| 401 |
+
"grad_norm": 636.6416015625,
|
| 402 |
+
"kl": 1.462548828125,
|
| 403 |
"learning_rate": 1.341705277779715e-06,
|
| 404 |
+
"loss": 0.0403,
|
| 405 |
+
"reward": 1.3687500536441803,
|
| 406 |
+
"reward_std": 0.5565730266273021,
|
| 407 |
+
"rewards/accuracy_reward": 0.5379464507102967,
|
| 408 |
+
"rewards/format_reward": 0.8308036103844643,
|
| 409 |
"step": 135
|
| 410 |
},
|
| 411 |
{
|
| 412 |
"clip_ratio": 0.0,
|
| 413 |
+
"completion_length": 473.5602897644043,
|
| 414 |
"epoch": 2.4093816631130065,
|
| 415 |
+
"grad_norm": 175.14183044433594,
|
| 416 |
+
"kl": 1.86015625,
|
| 417 |
"learning_rate": 1.2296174432791415e-06,
|
| 418 |
+
"loss": 0.0452,
|
| 419 |
+
"reward": 1.3828125596046448,
|
| 420 |
+
"reward_std": 0.5798827618360519,
|
| 421 |
+
"rewards/accuracy_reward": 0.5468750238418579,
|
| 422 |
+
"rewards/format_reward": 0.8359375432133674,
|
| 423 |
"step": 140
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"clip_ratio": 0.0,
|
| 427 |
+
"completion_length": 460.103589630127,
|
| 428 |
"epoch": 2.4946695095948828,
|
| 429 |
+
"grad_norm": 2649.130859375,
|
| 430 |
+
"kl": 2.87041015625,
|
| 431 |
"learning_rate": 1.1190709022599545e-06,
|
| 432 |
+
"loss": 0.0699,
|
| 433 |
+
"reward": 1.3917411416769028,
|
| 434 |
+
"reward_std": 0.5849474132061004,
|
| 435 |
+
"rewards/accuracy_reward": 0.556919663399458,
|
| 436 |
+
"rewards/format_reward": 0.8348214626312256,
|
| 437 |
"step": 145
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"clip_ratio": 0.0,
|
| 441 |
+
"completion_length": 458.2018081665039,
|
| 442 |
"epoch": 2.579957356076759,
|
| 443 |
+
"grad_norm": 1305.534423828125,
|
| 444 |
+
"kl": 4.1134765625,
|
| 445 |
"learning_rate": 1.0106958161686963e-06,
|
| 446 |
+
"loss": 0.1285,
|
| 447 |
+
"reward": 1.340178629755974,
|
| 448 |
+
"reward_std": 0.6000380210578442,
|
| 449 |
+
"rewards/accuracy_reward": 0.5174107395112515,
|
| 450 |
+
"rewards/format_reward": 0.8227678969502449,
|
| 451 |
"step": 150
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"clip_ratio": 0.0,
|
| 455 |
+
"completion_length": 452.27725372314455,
|
| 456 |
"epoch": 2.6652452025586353,
|
| 457 |
+
"grad_norm": 481.3914794921875,
|
| 458 |
+
"kl": 2.22734375,
|
| 459 |
"learning_rate": 9.051099682520474e-07,
|
| 460 |
+
"loss": 0.039,
|
| 461 |
+
"reward": 1.3504464954137803,
|
| 462 |
+
"reward_std": 0.613651292026043,
|
| 463 |
+
"rewards/accuracy_reward": 0.5243303805589676,
|
| 464 |
+
"rewards/format_reward": 0.826116107404232,
|
| 465 |
"step": 155
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"clip_ratio": 0.0,
|
| 469 |
+
"completion_length": 452.8314926147461,
|
| 470 |
"epoch": 2.750533049040512,
|
| 471 |
+
"grad_norm": 59.583213806152344,
|
| 472 |
+
"kl": 2.50859375,
|
| 473 |
"learning_rate": 8.029152419343472e-07,
|
| 474 |
+
"loss": 0.058,
|
| 475 |
+
"reward": 1.3649554133415223,
|
| 476 |
+
"reward_std": 0.6238026581704617,
|
| 477 |
+
"rewards/accuracy_reward": 0.544419664889574,
|
| 478 |
+
"rewards/format_reward": 0.8205357521772385,
|
| 479 |
"step": 160
|
| 480 |
},
|
| 481 |
{
|
| 482 |
"clip_ratio": 0.0,
|
| 483 |
+
"completion_length": 455.3404197692871,
|
| 484 |
"epoch": 2.835820895522388,
|
| 485 |
+
"grad_norm": 1071.0487060546875,
|
| 486 |
+
"kl": 2.58671875,
|
| 487 |
"learning_rate": 7.046941898307347e-07,
|
| 488 |
+
"loss": 0.076,
|
| 489 |
+
"reward": 1.340848270058632,
|
| 490 |
+
"reward_std": 0.6379699215292931,
|
| 491 |
+
"rewards/accuracy_reward": 0.5301339507102967,
|
| 492 |
+
"rewards/format_reward": 0.8107143223285675,
|
| 493 |
"step": 165
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"clip_ratio": 0.0,
|
| 497 |
+
"completion_length": 461.10560073852537,
|
| 498 |
"epoch": 2.9211087420042645,
|
| 499 |
+
"grad_norm": 869.7614135742188,
|
| 500 |
+
"kl": 2.18603515625,
|
| 501 |
"learning_rate": 6.11006712953975e-07,
|
| 502 |
+
"loss": 0.06,
|
| 503 |
+
"reward": 1.3625000655651092,
|
| 504 |
+
"reward_std": 0.605516166985035,
|
| 505 |
+
"rewards/accuracy_reward": 0.5419643118977546,
|
| 506 |
+
"rewards/format_reward": 0.8205357491970062,
|
| 507 |
"step": 170
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"clip_ratio": 0.0,
|
| 511 |
+
"completion_length": 465.46427154541016,
|
| 512 |
"epoch": 3.0170575692963753,
|
| 513 |
+
"grad_norm": 37.821475982666016,
|
| 514 |
+
"kl": 3.72734375,
|
| 515 |
"learning_rate": 5.223868690448817e-07,
|
| 516 |
+
"loss": 0.0881,
|
| 517 |
+
"reward": 1.3287946999073028,
|
| 518 |
+
"reward_std": 0.6144991792738438,
|
| 519 |
+
"rewards/accuracy_reward": 0.5147321581840515,
|
| 520 |
+
"rewards/format_reward": 0.8140625327825546,
|
| 521 |
"step": 175
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"clip_ratio": 0.0,
|
| 525 |
+
"completion_length": 473.2102905273438,
|
| 526 |
"epoch": 3.1023454157782515,
|
| 527 |
+
"grad_norm": 2058.541015625,
|
| 528 |
+
"kl": 3.3796875,
|
| 529 |
"learning_rate": 4.3933982822017883e-07,
|
| 530 |
+
"loss": 0.108,
|
| 531 |
+
"reward": 1.3754464894533158,
|
| 532 |
+
"reward_std": 0.6123204372823239,
|
| 533 |
+
"rewards/accuracy_reward": 0.5506696686148643,
|
| 534 |
+
"rewards/format_reward": 0.8247768208384514,
|
| 535 |
"step": 180
|
| 536 |
},
|
| 537 |
{
|
| 538 |
"clip_ratio": 0.0,
|
| 539 |
+
"completion_length": 475.0154228210449,
|
| 540 |
"epoch": 3.1876332622601278,
|
| 541 |
+
"grad_norm": 267681.28125,
|
| 542 |
+
"kl": 63.9951171875,
|
| 543 |
"learning_rate": 3.6233899329188115e-07,
|
| 544 |
+
"loss": 3.57,
|
| 545 |
+
"reward": 1.3428572058677672,
|
| 546 |
+
"reward_std": 0.6201678223907947,
|
| 547 |
+
"rewards/accuracy_reward": 0.5258928827941418,
|
| 548 |
+
"rewards/format_reward": 0.8169643193483352,
|
| 549 |
"step": 185
|
| 550 |
},
|
| 551 |
{
|
| 552 |
"clip_ratio": 0.0,
|
| 553 |
+
"completion_length": 471.1038162231445,
|
| 554 |
"epoch": 3.272921108742004,
|
| 555 |
+
"grad_norm": 2452.4267578125,
|
| 556 |
+
"kl": 32.8134765625,
|
| 557 |
"learning_rate": 2.9182330117358096e-07,
|
| 558 |
+
"loss": 1.8761,
|
| 559 |
+
"reward": 1.3464286357164383,
|
| 560 |
+
"reward_std": 0.6306888595223427,
|
| 561 |
+
"rewards/accuracy_reward": 0.5354910954833031,
|
| 562 |
+
"rewards/format_reward": 0.810937537252903,
|
| 563 |
"step": 190
|
| 564 |
},
|
| 565 |
{
|
| 566 |
"clip_ratio": 0.0,
|
| 567 |
+
"completion_length": 461.9361801147461,
|
| 568 |
"epoch": 3.3582089552238807,
|
| 569 |
+
"grad_norm": 938.4805908203125,
|
| 570 |
+
"kl": 4.35234375,
|
| 571 |
"learning_rate": 2.281947207567473e-07,
|
| 572 |
+
"loss": 0.131,
|
| 573 |
+
"reward": 1.3305804163217545,
|
| 574 |
+
"reward_std": 0.6262686759233475,
|
| 575 |
+
"rewards/accuracy_reward": 0.5127232395112514,
|
| 576 |
+
"rewards/format_reward": 0.8178571835160255,
|
| 577 |
"step": 195
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 3.443496801705757,
|
| 581 |
+
"grad_norm": 1083.9442138671875,
|
| 582 |
"learning_rate": 1.718159615201853e-07,
|
| 583 |
+
"loss": 0.0121,
|
| 584 |
"step": 200
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 3.443496801705757,
|
| 588 |
"eval_clip_ratio": 0.0,
|
| 589 |
+
"eval_completion_length": 467.76792581089006,
|
| 590 |
+
"eval_kl": 2.3580770766773163,
|
| 591 |
+
"eval_loss": 0.06606976687908173,
|
| 592 |
+
"eval_reward": 1.3115872357980893,
|
| 593 |
+
"eval_reward_std": 0.6067914157248915,
|
| 594 |
+
"eval_rewards/accuracy_reward": 0.492383637367346,
|
| 595 |
+
"eval_rewards/format_reward": 0.8192035969073018,
|
| 596 |
+
"eval_runtime": 4249.3378,
|
| 597 |
+
"eval_samples_per_second": 1.177,
|
| 598 |
"eval_steps_per_second": 0.011,
|
| 599 |
"step": 200
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"clip_ratio": 0.0,
|
| 603 |
+
"completion_length": 464.8625213623047,
|
| 604 |
"epoch": 3.5287846481876333,
|
| 605 |
+
"grad_norm": 41.21245193481445,
|
| 606 |
+
"kl": 2.200341796875,
|
| 607 |
"learning_rate": 1.2300840593454622e-07,
|
| 608 |
+
"loss": 0.076,
|
| 609 |
+
"reward": 1.3582589879631997,
|
| 610 |
+
"reward_std": 0.6224493138492108,
|
| 611 |
+
"rewards/accuracy_reward": 0.5436384178698063,
|
| 612 |
+
"rewards/format_reward": 0.8146205730736256,
|
| 613 |
"step": 205
|
| 614 |
},
|
| 615 |
{
|
| 616 |
"clip_ratio": 0.0,
|
| 617 |
+
"completion_length": 465.8446632385254,
|
| 618 |
"epoch": 3.6140724946695095,
|
| 619 |
+
"grad_norm": 418.15625,
|
| 620 |
+
"kl": 2.2728515625,
|
| 621 |
"learning_rate": 8.20502774480395e-08,
|
| 622 |
+
"loss": 0.0404,
|
| 623 |
+
"reward": 1.377901840209961,
|
| 624 |
+
"reward_std": 0.6090469680726528,
|
| 625 |
+
"rewards/accuracy_reward": 0.5529018096625805,
|
| 626 |
+
"rewards/format_reward": 0.8250000357627869,
|
| 627 |
"step": 210
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"clip_ratio": 0.0,
|
| 631 |
+
"completion_length": 464.74443969726565,
|
| 632 |
"epoch": 3.699360341151386,
|
| 633 |
+
"grad_norm": 315.2561340332031,
|
| 634 |
+
"kl": 2.88203125,
|
| 635 |
"learning_rate": 4.917505449659615e-08,
|
| 636 |
+
"loss": 0.0884,
|
| 637 |
+
"reward": 1.368080422282219,
|
| 638 |
+
"reward_std": 0.6142228864133358,
|
| 639 |
+
"rewards/accuracy_reward": 0.5479910932481289,
|
| 640 |
+
"rewards/format_reward": 0.8200893253087997,
|
| 641 |
"step": 215
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"clip_ratio": 0.0,
|
| 645 |
+
"completion_length": 467.38953018188477,
|
| 646 |
"epoch": 3.7846481876332625,
|
| 647 |
+
"grad_norm": 156.9326171875,
|
| 648 |
+
"kl": 2.8296875,
|
| 649 |
"learning_rate": 2.4570139579284723e-08,
|
| 650 |
+
"loss": 0.0658,
|
| 651 |
+
"reward": 1.3386161386966706,
|
| 652 |
+
"reward_std": 0.6271648786962032,
|
| 653 |
+
"rewards/accuracy_reward": 0.5265625193715096,
|
| 654 |
+
"rewards/format_reward": 0.8120536088943482,
|
| 655 |
"step": 220
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"clip_ratio": 0.0,
|
| 659 |
+
"completion_length": 462.0622955322266,
|
| 660 |
"epoch": 3.8699360341151388,
|
| 661 |
+
"grad_norm": 81.24732208251953,
|
| 662 |
+
"kl": 2.6775390625,
|
| 663 |
"learning_rate": 8.37579098581176e-09,
|
| 664 |
+
"loss": 0.0865,
|
| 665 |
+
"reward": 1.3390625447034836,
|
| 666 |
+
"reward_std": 0.6189317315816879,
|
| 667 |
+
"rewards/accuracy_reward": 0.5323660977184772,
|
| 668 |
+
"rewards/format_reward": 0.8066964671015739,
|
| 669 |
"step": 225
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"clip_ratio": 0.0,
|
| 673 |
+
"completion_length": 470.36943969726565,
|
| 674 |
"epoch": 3.955223880597015,
|
| 675 |
+
"grad_norm": 553.80859375,
|
| 676 |
+
"kl": 2.40078125,
|
| 677 |
"learning_rate": 6.843232656998933e-10,
|
| 678 |
+
"loss": 0.053,
|
| 679 |
+
"reward": 1.3424107700586319,
|
| 680 |
+
"reward_std": 0.6173162661492825,
|
| 681 |
+
"rewards/accuracy_reward": 0.5263393089175225,
|
| 682 |
+
"rewards/format_reward": 0.8160714641213417,
|
| 683 |
"step": 230
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"clip_ratio": 0.0,
|
| 687 |
+
"completion_length": 486.319974899292,
|
| 688 |
"epoch": 3.9893390191897655,
|
| 689 |
+
"kl": 2.2841796875,
|
| 690 |
+
"reward": 1.3577009439468384,
|
| 691 |
+
"reward_std": 0.5995111986994743,
|
| 692 |
+
"rewards/accuracy_reward": 0.5401785913854837,
|
| 693 |
+
"rewards/format_reward": 0.8175223655998707,
|
| 694 |
"step": 232,
|
| 695 |
"total_flos": 0.0,
|
| 696 |
+
"train_loss": 2.3046215271270145,
|
| 697 |
+
"train_runtime": 39674.2598,
|
| 698 |
+
"train_samples_per_second": 0.756,
|
| 699 |
"train_steps_per_second": 0.006
|
| 700 |
}
|
| 701 |
],
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7992
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b39e274769e16b9d724b7c7e093d76fc3e115e88841b6fcfd63b455cd60b72f8
|
| 3 |
size 7992
|