Training in progress, step 100, checkpoint
Browse files- checkpoint-100/model.safetensors +1 -1
- checkpoint-100/optimizer.pt +1 -1
- checkpoint-100/rng_state_0.pth +1 -1
- checkpoint-100/rng_state_1.pth +2 -2
- checkpoint-100/rng_state_2.pth +2 -2
- checkpoint-100/rng_state_3.pth +1 -1
- checkpoint-100/trainer_state.json +711 -711
- checkpoint-100/training_args.bin +2 -2
checkpoint-100/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1976163472
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad242d2f09d0c8055654d9bf3cc76ef68efe210b059d9b41e2e0bb07f5607f65
|
| 3 |
size 1976163472
|
checkpoint-100/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3952505274
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d18f3ee815fe1b06aa8c6b95f73354f741c1d7cb3d51e6010f8769c80ec5b6a
|
| 3 |
size 3952505274
|
checkpoint-100/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae69984fdd13c24018ecb1aa802b54afa3ca3e6a5ff1a82b51b4de545f616c19
|
| 3 |
size 15024
|
checkpoint-100/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cba81e50fb33c36c59d80bf8793f39dc24433a2190b46f8d2803c7db580c7ef5
|
| 3 |
+
size 14960
|
checkpoint-100/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f63266ab578a5d64086913db0f644de0a295ba0be0b8d58ecad8228d9505d57
|
| 3 |
+
size 14960
|
checkpoint-100/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:701c2a23e45bbff29fa1f6df5af29dcc2cdb24bd86078096f4fb5abd74ce9d23
|
| 3 |
size 15024
|
checkpoint-100/trainer_state.json
CHANGED
|
@@ -9,7 +9,7 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"completion_length":
|
| 13 |
"epoch": 0.0013333333333333333,
|
| 14 |
"grad_norm": 0.0,
|
| 15 |
"kl": 0.0,
|
|
@@ -22,1290 +22,1290 @@
|
|
| 22 |
"step": 1
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"completion_length":
|
| 26 |
"epoch": 0.0026666666666666666,
|
| 27 |
-
"grad_norm":
|
| 28 |
"kl": 0.0,
|
| 29 |
"learning_rate": 4.995066821070679e-07,
|
| 30 |
-
"loss":
|
| 31 |
-
"reward": 0.
|
| 32 |
-
"reward_std": 0.
|
| 33 |
-
"rewards/emotion_reward_func": 0.
|
| 34 |
-
"rewards/format_reward_func": 0.
|
| 35 |
"step": 2
|
| 36 |
},
|
| 37 |
{
|
| 38 |
-
"completion_length":
|
| 39 |
"epoch": 0.004,
|
| 40 |
-
"grad_norm":
|
| 41 |
-
"kl": 0.
|
| 42 |
"learning_rate": 4.9889049115077e-07,
|
| 43 |
"loss": 0.0,
|
| 44 |
-
"reward": 0.
|
| 45 |
-
"reward_std": 0.
|
| 46 |
-
"rewards/emotion_reward_func": 0.
|
| 47 |
-
"rewards/format_reward_func": 0.
|
| 48 |
"step": 3
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"completion_length":
|
| 52 |
"epoch": 0.005333333333333333,
|
| 53 |
-
"grad_norm":
|
| 54 |
-
"kl": 0.
|
| 55 |
"learning_rate": 4.980286753286194e-07,
|
| 56 |
"loss": 0.0,
|
| 57 |
-
"reward": 0.
|
| 58 |
-
"reward_std": 0.
|
| 59 |
-
"rewards/emotion_reward_func": 0.
|
| 60 |
-
"rewards/format_reward_func": 0.
|
| 61 |
"step": 4
|
| 62 |
},
|
| 63 |
{
|
| 64 |
-
"completion_length":
|
| 65 |
"epoch": 0.006666666666666667,
|
| 66 |
-
"grad_norm":
|
| 67 |
-
"kl": 0.
|
| 68 |
"learning_rate": 4.969220851487844e-07,
|
| 69 |
"loss": 0.0,
|
| 70 |
-
"reward": 0.
|
| 71 |
-
"reward_std": 0.
|
| 72 |
-
"rewards/emotion_reward_func": 0.
|
| 73 |
-
"rewards/format_reward_func": 0.
|
| 74 |
"step": 5
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"completion_length":
|
| 78 |
"epoch": 0.008,
|
| 79 |
-
"grad_norm":
|
| 80 |
-
"kl": 0.
|
| 81 |
"learning_rate": 4.955718126821722e-07,
|
| 82 |
"loss": 0.0,
|
| 83 |
-
"reward": 0.
|
| 84 |
-
"reward_std": 0.
|
| 85 |
-
"rewards/emotion_reward_func": 0.
|
| 86 |
-
"rewards/format_reward_func": 0.
|
| 87 |
"step": 6
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"completion_length":
|
| 91 |
"epoch": 0.009333333333333334,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"kl": 0.
|
| 94 |
"learning_rate": 4.939791904846868e-07,
|
| 95 |
"loss": 0.0,
|
| 96 |
-
"reward": 0.
|
| 97 |
"reward_std": 0.125,
|
| 98 |
-
"rewards/emotion_reward_func": 0.
|
| 99 |
-
"rewards/format_reward_func": 0.
|
| 100 |
"step": 7
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"completion_length":
|
| 104 |
"epoch": 0.010666666666666666,
|
| 105 |
-
"grad_norm":
|
| 106 |
-
"kl": 0.
|
| 107 |
"learning_rate": 4.921457902821578e-07,
|
| 108 |
"loss": 0.0,
|
| 109 |
-
"reward": 0.
|
| 110 |
-
"reward_std": 0.
|
| 111 |
-
"rewards/emotion_reward_func": 0.
|
| 112 |
-
"rewards/format_reward_func": 0.
|
| 113 |
"step": 8
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"completion_length":
|
| 117 |
"epoch": 0.012,
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"kl": 0.
|
| 120 |
"learning_rate": 4.900734214192358e-07,
|
| 121 |
"loss": 0.0,
|
| 122 |
-
"reward": 0.
|
| 123 |
-
"reward_std": 0.
|
| 124 |
-
"rewards/emotion_reward_func": 0.
|
| 125 |
-
"rewards/format_reward_func": 0.
|
| 126 |
"step": 9
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"completion_length":
|
| 130 |
"epoch": 0.013333333333333334,
|
| 131 |
-
"grad_norm": 6.
|
| 132 |
-
"kl": 0.
|
| 133 |
"learning_rate": 4.877641290737883e-07,
|
| 134 |
"loss": 0.0,
|
| 135 |
-
"reward": 0.
|
| 136 |
-
"reward_std": 0.
|
| 137 |
-
"rewards/emotion_reward_func": 0.
|
| 138 |
-
"rewards/format_reward_func": 0.
|
| 139 |
"step": 10
|
| 140 |
},
|
| 141 |
{
|
| 142 |
-
"completion_length":
|
| 143 |
"epoch": 0.014666666666666666,
|
| 144 |
-
"grad_norm": 6.
|
| 145 |
-
"kl": 0.
|
| 146 |
"learning_rate": 4.852201922385564e-07,
|
| 147 |
"loss": 0.0,
|
| 148 |
-
"reward": 0.
|
| 149 |
-
"reward_std": 0.
|
| 150 |
-
"rewards/emotion_reward_func": 0.
|
| 151 |
-
"rewards/format_reward_func": 0.
|
| 152 |
"step": 11
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"completion_length":
|
| 156 |
"epoch": 0.016,
|
| 157 |
-
"grad_norm":
|
| 158 |
-
"kl": 0.
|
| 159 |
"learning_rate": 4.824441214720628e-07,
|
| 160 |
"loss": 0.0,
|
| 161 |
-
"reward": 0.
|
| 162 |
-
"reward_std": 0.
|
| 163 |
-
"rewards/emotion_reward_func": 0.
|
| 164 |
-
"rewards/format_reward_func": 0.
|
| 165 |
"step": 12
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"completion_length":
|
| 169 |
"epoch": 0.017333333333333333,
|
| 170 |
-
"grad_norm":
|
| 171 |
-
"kl": 0.
|
| 172 |
"learning_rate": 4.794386564209952e-07,
|
| 173 |
"loss": 0.0,
|
| 174 |
-
"reward": 0.
|
| 175 |
-
"reward_std": 0.
|
| 176 |
-
"rewards/emotion_reward_func": 0.
|
| 177 |
-
"rewards/format_reward_func": 0.
|
| 178 |
"step": 13
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"completion_length":
|
| 182 |
"epoch": 0.018666666666666668,
|
| 183 |
-
"grad_norm":
|
| 184 |
-
"kl": 0.
|
| 185 |
"learning_rate": 4.762067631165049e-07,
|
| 186 |
"loss": 0.0,
|
| 187 |
-
"reward": 0.
|
| 188 |
-
"reward_std": 0.
|
| 189 |
-
"rewards/emotion_reward_func": 0.
|
| 190 |
-
"rewards/format_reward_func": 0.
|
| 191 |
"step": 14
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"completion_length":
|
| 195 |
"epoch": 0.02,
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"kl": 0.
|
| 198 |
"learning_rate": 4.7275163104709194e-07,
|
| 199 |
-
"loss": 0.
|
| 200 |
-
"reward": 0.
|
| 201 |
-
"reward_std": 0.
|
| 202 |
-
"rewards/emotion_reward_func": 0.
|
| 203 |
-
"rewards/format_reward_func": 0.
|
| 204 |
"step": 15
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"completion_length":
|
| 208 |
"epoch": 0.021333333333333333,
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"kl": 0.
|
| 211 |
"learning_rate": 4.6907667001096585e-07,
|
| 212 |
"loss": 0.0,
|
| 213 |
-
"reward": 0.
|
| 214 |
-
"reward_std": 0.
|
| 215 |
-
"rewards/emotion_reward_func": 0.
|
| 216 |
-
"rewards/format_reward_func": 0.
|
| 217 |
"step": 16
|
| 218 |
},
|
| 219 |
{
|
| 220 |
-
"completion_length":
|
| 221 |
"epoch": 0.02266666666666667,
|
| 222 |
-
"grad_norm":
|
| 223 |
-
"kl": 0.
|
| 224 |
"learning_rate": 4.6518550675098587e-07,
|
| 225 |
"loss": 0.0,
|
| 226 |
-
"reward": 0.
|
| 227 |
-
"reward_std": 0.
|
| 228 |
-
"rewards/emotion_reward_func": 0.
|
| 229 |
-
"rewards/format_reward_func": 0.
|
| 230 |
"step": 17
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"completion_length":
|
| 234 |
"epoch": 0.024,
|
| 235 |
-
"grad_norm":
|
| 236 |
-
"kl": 0.
|
| 237 |
"learning_rate": 4.6108198137550377e-07,
|
| 238 |
"loss": 0.0,
|
| 239 |
-
"reward": 0.
|
| 240 |
-
"reward_std": 0.
|
| 241 |
-
"rewards/emotion_reward_func": 0.
|
| 242 |
-
"rewards/format_reward_func": 0.
|
| 243 |
"step": 18
|
| 244 |
},
|
| 245 |
{
|
| 246 |
-
"completion_length":
|
| 247 |
"epoch": 0.025333333333333333,
|
| 248 |
-
"grad_norm":
|
| 249 |
-
"kl": 0.
|
| 250 |
"learning_rate": 4.567701435686404e-07,
|
| 251 |
"loss": 0.0,
|
| 252 |
-
"reward": 0.
|
| 253 |
-
"reward_std": 0.
|
| 254 |
-
"rewards/emotion_reward_func": 0.
|
| 255 |
-
"rewards/format_reward_func": 0.
|
| 256 |
"step": 19
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"completion_length":
|
| 260 |
"epoch": 0.02666666666666667,
|
| 261 |
-
"grad_norm": 8.
|
| 262 |
-
"kl": 0.
|
| 263 |
"learning_rate": 4.5225424859373684e-07,
|
| 264 |
"loss": 0.0,
|
| 265 |
-
"reward": 0.
|
| 266 |
-
"reward_std": 0.
|
| 267 |
-
"rewards/emotion_reward_func": 0.
|
| 268 |
-
"rewards/format_reward_func": 0.
|
| 269 |
"step": 20
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"completion_length":
|
| 273 |
"epoch": 0.028,
|
| 274 |
-
"grad_norm":
|
| 275 |
-
"kl": 0.
|
| 276 |
"learning_rate": 4.475387530939226e-07,
|
| 277 |
"loss": 0.0,
|
| 278 |
-
"reward": 0.
|
| 279 |
-
"reward_std": 0.
|
| 280 |
-
"rewards/emotion_reward_func": 0.
|
| 281 |
-
"rewards/format_reward_func": 0.
|
| 282 |
"step": 21
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"completion_length":
|
| 286 |
"epoch": 0.029333333333333333,
|
| 287 |
-
"grad_norm":
|
| 288 |
-
"kl": 0.
|
| 289 |
"learning_rate": 4.426283106939473e-07,
|
| 290 |
"loss": 0.0,
|
| 291 |
-
"reward":
|
| 292 |
-
"reward_std": 0.
|
| 293 |
-
"rewards/emotion_reward_func": 0.
|
| 294 |
-
"rewards/format_reward_func": 0.
|
| 295 |
"step": 22
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"completion_length":
|
| 299 |
"epoch": 0.030666666666666665,
|
| 300 |
-
"grad_norm":
|
| 301 |
-
"kl": 0.
|
| 302 |
"learning_rate": 4.375277674076149e-07,
|
| 303 |
"loss": 0.0,
|
| 304 |
-
"reward":
|
| 305 |
-
"reward_std": 0.
|
| 306 |
-
"rewards/emotion_reward_func": 0.
|
| 307 |
-
"rewards/format_reward_func": 0.
|
| 308 |
"step": 23
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"completion_length":
|
| 312 |
"epoch": 0.032,
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"kl": 0.
|
| 315 |
"learning_rate": 4.3224215685535287e-07,
|
| 316 |
-
"loss": 0.
|
| 317 |
-
"reward":
|
| 318 |
-
"reward_std": 0.
|
| 319 |
-
"rewards/emotion_reward_func": 0.
|
| 320 |
-
"rewards/format_reward_func": 0.
|
| 321 |
"step": 24
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"completion_length":
|
| 325 |
"epoch": 0.03333333333333333,
|
| 326 |
-
"grad_norm":
|
| 327 |
-
"kl": 0.
|
| 328 |
"learning_rate": 4.2677669529663686e-07,
|
| 329 |
-
"loss": 0.
|
| 330 |
-
"reward": 0.
|
| 331 |
-
"reward_std": 0.
|
| 332 |
-
"rewards/emotion_reward_func": 0.
|
| 333 |
-
"rewards/format_reward_func": 0.
|
| 334 |
"step": 25
|
| 335 |
},
|
| 336 |
{
|
| 337 |
-
"completion_length":
|
| 338 |
"epoch": 0.034666666666666665,
|
| 339 |
-
"grad_norm":
|
| 340 |
-
"kl": 0.
|
| 341 |
"learning_rate": 4.2113677648217216e-07,
|
| 342 |
"loss": 0.0001,
|
| 343 |
-
"reward":
|
| 344 |
-
"reward_std": 0.
|
| 345 |
-
"rewards/emotion_reward_func": 0.
|
| 346 |
-
"rewards/format_reward_func": 0.
|
| 347 |
"step": 26
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"completion_length":
|
| 351 |
"epoch": 0.036,
|
| 352 |
-
"grad_norm":
|
| 353 |
-
"kl": 0.
|
| 354 |
"learning_rate": 4.1532796633091294e-07,
|
| 355 |
-
"loss": 0.
|
| 356 |
-
"reward": 0
|
| 357 |
-
"reward_std": 0.
|
| 358 |
-
"rewards/emotion_reward_func": 0.
|
| 359 |
-
"rewards/format_reward_func": 0.
|
| 360 |
"step": 27
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"completion_length":
|
| 364 |
"epoch": 0.037333333333333336,
|
| 365 |
-
"grad_norm":
|
| 366 |
-
"kl": 0.
|
| 367 |
"learning_rate": 4.0935599743717244e-07,
|
| 368 |
"loss": 0.0001,
|
| 369 |
-
"reward":
|
| 370 |
-
"reward_std": 0.
|
| 371 |
-
"rewards/emotion_reward_func": 0.
|
| 372 |
-
"rewards/format_reward_func": 0.
|
| 373 |
"step": 28
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"completion_length":
|
| 377 |
"epoch": 0.03866666666666667,
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"kl": 0.
|
| 380 |
"learning_rate": 4.0322676341324414e-07,
|
| 381 |
-
"loss": 0.
|
| 382 |
-
"reward":
|
| 383 |
-
"reward_std": 0.
|
| 384 |
-
"rewards/emotion_reward_func": 0.
|
| 385 |
-
"rewards/format_reward_func": 0.
|
| 386 |
"step": 29
|
| 387 |
},
|
| 388 |
{
|
| 389 |
-
"completion_length":
|
| 390 |
"epoch": 0.04,
|
| 391 |
-
"grad_norm":
|
| 392 |
-
"kl": 0.
|
| 393 |
"learning_rate": 3.9694631307311825e-07,
|
| 394 |
-
"loss": 0.
|
| 395 |
-
"reward":
|
| 396 |
-
"reward_std": 0.
|
| 397 |
-
"rewards/emotion_reward_func": 0.
|
| 398 |
-
"rewards/format_reward_func": 0.
|
| 399 |
"step": 30
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"completion_length":
|
| 403 |
"epoch": 0.04133333333333333,
|
| 404 |
-
"grad_norm":
|
| 405 |
-
"kl": 0.
|
| 406 |
"learning_rate": 3.9052084446303265e-07,
|
| 407 |
"loss": 0.0001,
|
| 408 |
-
"reward":
|
| 409 |
-
"reward_std": 0.
|
| 410 |
-
"rewards/emotion_reward_func": 0.
|
| 411 |
-
"rewards/format_reward_func": 0.
|
| 412 |
"step": 31
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"completion_length":
|
| 416 |
"epoch": 0.042666666666666665,
|
| 417 |
-
"grad_norm":
|
| 418 |
-
"kl": 0.
|
| 419 |
"learning_rate": 3.839566987447491e-07,
|
| 420 |
-
"loss": 0.
|
| 421 |
-
"reward":
|
| 422 |
-
"reward_std": 0.
|
| 423 |
-
"rewards/emotion_reward_func": 0.
|
| 424 |
-
"rewards/format_reward_func": 0.
|
| 425 |
"step": 32
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"completion_length":
|
| 429 |
"epoch": 0.044,
|
| 430 |
-
"grad_norm":
|
| 431 |
-
"kl": 0.
|
| 432 |
"learning_rate": 3.7726035393759283e-07,
|
| 433 |
"loss": 0.0001,
|
| 434 |
-
"reward":
|
| 435 |
-
"reward_std": 0.
|
| 436 |
-
"rewards/emotion_reward_func": 0.
|
| 437 |
-
"rewards/format_reward_func": 0.
|
| 438 |
"step": 33
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"completion_length":
|
| 442 |
"epoch": 0.04533333333333334,
|
| 443 |
-
"grad_norm": 6.
|
| 444 |
-
"kl": 0.
|
| 445 |
"learning_rate": 3.704384185254288e-07,
|
| 446 |
"loss": 0.0001,
|
| 447 |
-
"reward":
|
| 448 |
-
"reward_std": 0.
|
| 449 |
-
"rewards/emotion_reward_func": 0.
|
| 450 |
-
"rewards/format_reward_func": 0.
|
| 451 |
"step": 34
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"completion_length":
|
| 455 |
"epoch": 0.04666666666666667,
|
| 456 |
-
"grad_norm":
|
| 457 |
-
"kl": 0.
|
| 458 |
"learning_rate": 3.634976249348867e-07,
|
| 459 |
"loss": 0.0001,
|
| 460 |
-
"reward":
|
| 461 |
-
"reward_std": 0.
|
| 462 |
-
"rewards/emotion_reward_func": 0.
|
| 463 |
-
"rewards/format_reward_func": 0.
|
| 464 |
"step": 35
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"completion_length":
|
| 468 |
"epoch": 0.048,
|
| 469 |
-
"grad_norm":
|
| 470 |
-
"kl": 0.
|
| 471 |
"learning_rate": 3.5644482289126813e-07,
|
| 472 |
"loss": 0.0001,
|
| 473 |
-
"reward":
|
| 474 |
-
"reward_std": 0.
|
| 475 |
-
"rewards/emotion_reward_func": 0.
|
| 476 |
-
"rewards/format_reward_func": 0.
|
| 477 |
"step": 36
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"completion_length":
|
| 481 |
"epoch": 0.04933333333333333,
|
| 482 |
-
"grad_norm":
|
| 483 |
-
"kl": 0.
|
| 484 |
"learning_rate": 3.492869726586951e-07,
|
| 485 |
"loss": 0.0001,
|
| 486 |
-
"reward":
|
| 487 |
-
"reward_std": 0.
|
| 488 |
-
"rewards/emotion_reward_func": 0.
|
| 489 |
-
"rewards/format_reward_func": 0.
|
| 490 |
"step": 37
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"completion_length":
|
| 494 |
"epoch": 0.050666666666666665,
|
| 495 |
-
"grad_norm":
|
| 496 |
-
"kl": 0.
|
| 497 |
"learning_rate": 3.4203113817116953e-07,
|
| 498 |
"loss": 0.0001,
|
| 499 |
-
"reward":
|
| 500 |
-
"reward_std": 0.
|
| 501 |
-
"rewards/emotion_reward_func": 0.
|
| 502 |
-
"rewards/format_reward_func": 0.
|
| 503 |
"step": 38
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"completion_length":
|
| 507 |
"epoch": 0.052,
|
| 508 |
-
"grad_norm":
|
| 509 |
-
"kl": 0.
|
| 510 |
"learning_rate": 3.346844800613229e-07,
|
| 511 |
"loss": 0.0001,
|
| 512 |
-
"reward":
|
| 513 |
-
"reward_std": 0.
|
| 514 |
-
"rewards/emotion_reward_func": 0.
|
| 515 |
-
"rewards/format_reward_func": 0.
|
| 516 |
"step": 39
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"completion_length":
|
| 520 |
"epoch": 0.05333333333333334,
|
| 521 |
-
"grad_norm":
|
| 522 |
-
"kl": 0.
|
| 523 |
"learning_rate": 3.272542485937368e-07,
|
| 524 |
"loss": 0.0001,
|
| 525 |
-
"reward":
|
| 526 |
-
"reward_std": 0.
|
| 527 |
-
"rewards/emotion_reward_func": 0.
|
| 528 |
-
"rewards/format_reward_func": 0.
|
| 529 |
"step": 40
|
| 530 |
},
|
| 531 |
{
|
| 532 |
-
"completion_length":
|
| 533 |
"epoch": 0.05466666666666667,
|
| 534 |
-
"grad_norm":
|
| 535 |
-
"kl": 0.
|
| 536 |
"learning_rate": 3.1974777650980734e-07,
|
| 537 |
"loss": 0.0001,
|
| 538 |
-
"reward":
|
| 539 |
-
"reward_std": 0.
|
| 540 |
-
"rewards/emotion_reward_func": 0.
|
| 541 |
-
"rewards/format_reward_func": 0.
|
| 542 |
"step": 41
|
| 543 |
},
|
| 544 |
{
|
| 545 |
-
"completion_length":
|
| 546 |
"epoch": 0.056,
|
| 547 |
-
"grad_norm":
|
| 548 |
-
"kl": 0.
|
| 549 |
"learning_rate": 3.121724717912138e-07,
|
| 550 |
"loss": 0.0001,
|
| 551 |
-
"reward":
|
| 552 |
-
"reward_std": 0.
|
| 553 |
-
"rewards/emotion_reward_func": 0.
|
| 554 |
-
"rewards/format_reward_func": 0
|
| 555 |
"step": 42
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"completion_length":
|
| 559 |
"epoch": 0.05733333333333333,
|
| 560 |
-
"grad_norm":
|
| 561 |
-
"kl": 0.
|
| 562 |
"learning_rate": 3.0453581034913565e-07,
|
| 563 |
"loss": 0.0001,
|
| 564 |
-
"reward":
|
| 565 |
-
"reward_std": 0.
|
| 566 |
-
"rewards/emotion_reward_func": 0.
|
| 567 |
-
"rewards/format_reward_func": 0.
|
| 568 |
"step": 43
|
| 569 |
},
|
| 570 |
{
|
| 571 |
-
"completion_length":
|
| 572 |
"epoch": 0.058666666666666666,
|
| 573 |
-
"grad_norm":
|
| 574 |
-
"kl": 0.
|
| 575 |
"learning_rate": 2.968453286464312e-07,
|
| 576 |
"loss": 0.0001,
|
| 577 |
-
"reward":
|
| 578 |
-
"reward_std": 0.
|
| 579 |
-
"rewards/emotion_reward_func": 0.
|
| 580 |
-
"rewards/format_reward_func": 0.
|
| 581 |
"step": 44
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"completion_length":
|
| 585 |
"epoch": 0.06,
|
| 586 |
-
"grad_norm":
|
| 587 |
-
"kl": 0.
|
| 588 |
"learning_rate": 2.8910861626005773e-07,
|
| 589 |
"loss": 0.0001,
|
| 590 |
-
"reward":
|
| 591 |
-
"reward_std": 0.
|
| 592 |
-
"rewards/emotion_reward_func": 0.
|
| 593 |
-
"rewards/format_reward_func": 0.
|
| 594 |
"step": 45
|
| 595 |
},
|
| 596 |
{
|
| 597 |
-
"completion_length":
|
| 598 |
"epoch": 0.06133333333333333,
|
| 599 |
-
"grad_norm": 6.
|
| 600 |
-
"kl": 0.
|
| 601 |
"learning_rate": 2.8133330839107604e-07,
|
| 602 |
"loss": 0.0001,
|
| 603 |
-
"reward":
|
| 604 |
-
"reward_std": 0.
|
| 605 |
-
"rewards/emotion_reward_func": 0.
|
| 606 |
-
"rewards/format_reward_func": 0.
|
| 607 |
"step": 46
|
| 608 |
},
|
| 609 |
{
|
| 610 |
-
"completion_length":
|
| 611 |
"epoch": 0.06266666666666666,
|
| 612 |
-
"grad_norm":
|
| 613 |
-
"kl": 0.
|
| 614 |
"learning_rate": 2.735270783296286e-07,
|
| 615 |
"loss": 0.0001,
|
| 616 |
-
"reward":
|
| 617 |
-
"reward_std": 0.
|
| 618 |
-
"rewards/emotion_reward_func": 0.
|
| 619 |
-
"rewards/format_reward_func": 0.
|
| 620 |
"step": 47
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"completion_length":
|
| 624 |
"epoch": 0.064,
|
| 625 |
-
"grad_norm":
|
| 626 |
-
"kl": 0.
|
| 627 |
"learning_rate": 2.6569762988232837e-07,
|
| 628 |
"loss": 0.0001,
|
| 629 |
-
"reward":
|
| 630 |
-
"reward_std": 0.
|
| 631 |
-
"rewards/emotion_reward_func": 0.
|
| 632 |
-
"rewards/format_reward_func": 0.
|
| 633 |
"step": 48
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"completion_length":
|
| 637 |
"epoch": 0.06533333333333333,
|
| 638 |
-
"grad_norm":
|
| 639 |
-
"kl": 0.
|
| 640 |
"learning_rate": 2.5785268976953204e-07,
|
| 641 |
"loss": 0.0001,
|
| 642 |
-
"reward":
|
| 643 |
-
"reward_std": 0.
|
| 644 |
-
"rewards/emotion_reward_func": 0.
|
| 645 |
-
"rewards/format_reward_func": 0.
|
| 646 |
"step": 49
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"completion_length":
|
| 650 |
"epoch": 0.06666666666666667,
|
| 651 |
-
"grad_norm":
|
| 652 |
-
"kl": 0.
|
| 653 |
"learning_rate": 2.5e-07,
|
| 654 |
"loss": 0.0001,
|
| 655 |
-
"reward":
|
| 656 |
-
"reward_std": 0.
|
| 657 |
-
"rewards/emotion_reward_func": 0.
|
| 658 |
-
"rewards/format_reward_func": 0.
|
| 659 |
"step": 50
|
| 660 |
},
|
| 661 |
{
|
| 662 |
-
"completion_length":
|
| 663 |
"epoch": 0.068,
|
| 664 |
-
"grad_norm":
|
| 665 |
-
"kl": 0.
|
| 666 |
"learning_rate": 2.4214731023046794e-07,
|
| 667 |
"loss": 0.0001,
|
| 668 |
-
"reward":
|
| 669 |
-
"reward_std": 0.
|
| 670 |
-
"rewards/emotion_reward_func": 0.
|
| 671 |
-
"rewards/format_reward_func": 0.
|
| 672 |
"step": 51
|
| 673 |
},
|
| 674 |
{
|
| 675 |
-
"completion_length":
|
| 676 |
"epoch": 0.06933333333333333,
|
| 677 |
-
"grad_norm":
|
| 678 |
-
"kl": 0.
|
| 679 |
"learning_rate": 2.3430237011767164e-07,
|
| 680 |
"loss": 0.0001,
|
| 681 |
-
"reward":
|
| 682 |
-
"reward_std": 0.
|
| 683 |
-
"rewards/emotion_reward_func": 0.
|
| 684 |
-
"rewards/format_reward_func": 0
|
| 685 |
"step": 52
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"completion_length":
|
| 689 |
"epoch": 0.07066666666666667,
|
| 690 |
-
"grad_norm":
|
| 691 |
-
"kl": 0.
|
| 692 |
"learning_rate": 2.264729216703714e-07,
|
| 693 |
"loss": 0.0001,
|
| 694 |
-
"reward":
|
| 695 |
-
"reward_std": 0.
|
| 696 |
-
"rewards/emotion_reward_func": 0.
|
| 697 |
-
"rewards/format_reward_func": 0.
|
| 698 |
"step": 53
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"completion_length":
|
| 702 |
"epoch": 0.072,
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"kl": 0.
|
| 705 |
"learning_rate": 2.1866669160892389e-07,
|
| 706 |
"loss": 0.0001,
|
| 707 |
-
"reward": 1.
|
| 708 |
-
"reward_std": 0.
|
| 709 |
-
"rewards/emotion_reward_func": 0.
|
| 710 |
-
"rewards/format_reward_func":
|
| 711 |
"step": 54
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"completion_length":
|
| 715 |
"epoch": 0.07333333333333333,
|
| 716 |
-
"grad_norm":
|
| 717 |
-
"kl": 0.
|
| 718 |
"learning_rate": 2.1089138373994222e-07,
|
| 719 |
-
"loss": 0.
|
| 720 |
-
"reward":
|
| 721 |
-
"reward_std": 0.
|
| 722 |
-
"rewards/emotion_reward_func": 0.
|
| 723 |
-
"rewards/format_reward_func": 0.
|
| 724 |
"step": 55
|
| 725 |
},
|
| 726 |
{
|
| 727 |
-
"completion_length":
|
| 728 |
"epoch": 0.07466666666666667,
|
| 729 |
-
"grad_norm":
|
| 730 |
-
"kl": 0.
|
| 731 |
"learning_rate": 2.0315467135356878e-07,
|
| 732 |
"loss": 0.0001,
|
| 733 |
-
"reward":
|
| 734 |
-
"reward_std": 0.
|
| 735 |
-
"rewards/emotion_reward_func": 0.
|
| 736 |
-
"rewards/format_reward_func": 0
|
| 737 |
"step": 56
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"completion_length":
|
| 741 |
"epoch": 0.076,
|
| 742 |
-
"grad_norm":
|
| 743 |
-
"kl": 0.
|
| 744 |
"learning_rate": 1.954641896508644e-07,
|
| 745 |
-
"loss": 0.
|
| 746 |
-
"reward": 1.
|
| 747 |
-
"reward_std": 0.
|
| 748 |
-
"rewards/emotion_reward_func": 0.
|
| 749 |
-
"rewards/format_reward_func":
|
| 750 |
"step": 57
|
| 751 |
},
|
| 752 |
{
|
| 753 |
-
"completion_length":
|
| 754 |
"epoch": 0.07733333333333334,
|
| 755 |
-
"grad_norm":
|
| 756 |
-
"kl": 0.
|
| 757 |
"learning_rate": 1.8782752820878633e-07,
|
| 758 |
-
"loss": 0.
|
| 759 |
-
"reward":
|
| 760 |
-
"reward_std": 0.
|
| 761 |
-
"rewards/emotion_reward_func": 0.
|
| 762 |
-
"rewards/format_reward_func": 0.
|
| 763 |
"step": 58
|
| 764 |
},
|
| 765 |
{
|
| 766 |
-
"completion_length":
|
| 767 |
"epoch": 0.07866666666666666,
|
| 768 |
-
"grad_norm":
|
| 769 |
-
"kl": 0.
|
| 770 |
"learning_rate": 1.802522234901927e-07,
|
| 771 |
"loss": 0.0002,
|
| 772 |
-
"reward":
|
| 773 |
-
"reward_std": 0.
|
| 774 |
-
"rewards/emotion_reward_func": 0.
|
| 775 |
-
"rewards/format_reward_func": 0.
|
| 776 |
"step": 59
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"completion_length":
|
| 780 |
"epoch": 0.08,
|
| 781 |
-
"grad_norm":
|
| 782 |
-
"kl": 0.
|
| 783 |
"learning_rate": 1.7274575140626315e-07,
|
| 784 |
"loss": 0.0001,
|
| 785 |
-
"reward": 1.
|
| 786 |
-
"reward_std": 0.
|
| 787 |
-
"rewards/emotion_reward_func": 0.
|
| 788 |
-
"rewards/format_reward_func":
|
| 789 |
"step": 60
|
| 790 |
},
|
| 791 |
{
|
| 792 |
-
"completion_length":
|
| 793 |
"epoch": 0.08133333333333333,
|
| 794 |
-
"grad_norm":
|
| 795 |
-
"kl": 0.
|
| 796 |
"learning_rate": 1.6531551993867715e-07,
|
| 797 |
-
"loss": 0.
|
| 798 |
-
"reward":
|
| 799 |
-
"reward_std": 0.
|
| 800 |
-
"rewards/emotion_reward_func": 0.
|
| 801 |
-
"rewards/format_reward_func": 0.
|
| 802 |
"step": 61
|
| 803 |
},
|
| 804 |
{
|
| 805 |
-
"completion_length":
|
| 806 |
"epoch": 0.08266666666666667,
|
| 807 |
-
"grad_norm":
|
| 808 |
-
"kl": 0.
|
| 809 |
"learning_rate": 1.579688618288305e-07,
|
| 810 |
"loss": 0.0001,
|
| 811 |
-
"reward": 1.
|
| 812 |
-
"reward_std": 0.
|
| 813 |
-
"rewards/emotion_reward_func": 0.
|
| 814 |
"rewards/format_reward_func": 1.0,
|
| 815 |
"step": 62
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"completion_length":
|
| 819 |
"epoch": 0.084,
|
| 820 |
-
"grad_norm":
|
| 821 |
-
"kl": 0.
|
| 822 |
"learning_rate": 1.5071302734130486e-07,
|
| 823 |
-
"loss": 0.
|
| 824 |
-
"reward":
|
| 825 |
-
"reward_std": 0.
|
| 826 |
-
"rewards/emotion_reward_func": 0.
|
| 827 |
-
"rewards/format_reward_func": 0.
|
| 828 |
"step": 63
|
| 829 |
},
|
| 830 |
{
|
| 831 |
-
"completion_length":
|
| 832 |
"epoch": 0.08533333333333333,
|
| 833 |
-
"grad_norm":
|
| 834 |
-
"kl": 0.
|
| 835 |
"learning_rate": 1.4355517710873182e-07,
|
| 836 |
"loss": 0.0001,
|
| 837 |
-
"reward": 1.
|
| 838 |
-
"reward_std": 0.
|
| 839 |
-
"rewards/emotion_reward_func": 0.
|
| 840 |
-
"rewards/format_reward_func":
|
| 841 |
"step": 64
|
| 842 |
},
|
| 843 |
{
|
| 844 |
-
"completion_length":
|
| 845 |
"epoch": 0.08666666666666667,
|
| 846 |
-
"grad_norm":
|
| 847 |
-
"kl": 0.
|
| 848 |
"learning_rate": 1.365023750651133e-07,
|
| 849 |
"loss": 0.0001,
|
| 850 |
-
"reward":
|
| 851 |
-
"reward_std": 0.
|
| 852 |
-
"rewards/emotion_reward_func": 0.
|
| 853 |
-
"rewards/format_reward_func": 0.
|
| 854 |
"step": 65
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"completion_length":
|
| 858 |
"epoch": 0.088,
|
| 859 |
-
"grad_norm":
|
| 860 |
-
"kl": 0.
|
| 861 |
"learning_rate": 1.2956158147457114e-07,
|
| 862 |
"loss": 0.0001,
|
| 863 |
-
"reward":
|
| 864 |
-
"reward_std": 0.
|
| 865 |
-
"rewards/emotion_reward_func": 0.
|
| 866 |
-
"rewards/format_reward_func": 0.
|
| 867 |
"step": 66
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"completion_length":
|
| 871 |
"epoch": 0.08933333333333333,
|
| 872 |
-
"grad_norm":
|
| 873 |
-
"kl": 0.
|
| 874 |
"learning_rate": 1.2273964606240718e-07,
|
| 875 |
"loss": 0.0001,
|
| 876 |
-
"reward":
|
| 877 |
-
"reward_std": 0.
|
| 878 |
-
"rewards/emotion_reward_func": 0.
|
| 879 |
-
"rewards/format_reward_func": 0
|
| 880 |
"step": 67
|
| 881 |
},
|
| 882 |
{
|
| 883 |
-
"completion_length":
|
| 884 |
"epoch": 0.09066666666666667,
|
| 885 |
-
"grad_norm":
|
| 886 |
-
"kl": 0.
|
| 887 |
"learning_rate": 1.1604330125525078e-07,
|
| 888 |
"loss": 0.0001,
|
| 889 |
-
"reward": 1.
|
| 890 |
-
"reward_std": 0.
|
| 891 |
-
"rewards/emotion_reward_func": 0.
|
| 892 |
-
"rewards/format_reward_func":
|
| 893 |
"step": 68
|
| 894 |
},
|
| 895 |
{
|
| 896 |
-
"completion_length":
|
| 897 |
"epoch": 0.092,
|
| 898 |
-
"grad_norm":
|
| 899 |
-
"kl": 0.
|
| 900 |
"learning_rate": 1.0947915553696741e-07,
|
| 901 |
"loss": 0.0001,
|
| 902 |
-
"reward":
|
| 903 |
-
"reward_std": 0.
|
| 904 |
-
"rewards/emotion_reward_func": 0.
|
| 905 |
-
"rewards/format_reward_func": 0
|
| 906 |
"step": 69
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"completion_length":
|
| 910 |
"epoch": 0.09333333333333334,
|
| 911 |
-
"grad_norm":
|
| 912 |
-
"kl":
|
| 913 |
"learning_rate": 1.0305368692688174e-07,
|
| 914 |
-
"loss": 0.
|
| 915 |
-
"reward":
|
| 916 |
-
"reward_std": 0.
|
| 917 |
-
"rewards/emotion_reward_func": 0.
|
| 918 |
-
"rewards/format_reward_func": 0.
|
| 919 |
"step": 70
|
| 920 |
},
|
| 921 |
{
|
| 922 |
-
"completion_length":
|
| 923 |
"epoch": 0.09466666666666666,
|
| 924 |
-
"grad_norm":
|
| 925 |
-
"kl": 0.
|
| 926 |
"learning_rate": 9.677323658675593e-08,
|
| 927 |
"loss": 0.0001,
|
| 928 |
-
"reward":
|
| 929 |
-
"reward_std": 0.
|
| 930 |
-
"rewards/emotion_reward_func": 0.
|
| 931 |
-
"rewards/format_reward_func": 0.
|
| 932 |
"step": 71
|
| 933 |
},
|
| 934 |
{
|
| 935 |
-
"completion_length":
|
| 936 |
"epoch": 0.096,
|
| 937 |
-
"grad_norm": 5.
|
| 938 |
-
"kl": 0.
|
| 939 |
"learning_rate": 9.064400256282755e-08,
|
| 940 |
-
"loss": 0.
|
| 941 |
-
"reward":
|
| 942 |
-
"reward_std": 0.
|
| 943 |
-
"rewards/emotion_reward_func": 0.
|
| 944 |
-
"rewards/format_reward_func": 0.
|
| 945 |
"step": 72
|
| 946 |
},
|
| 947 |
{
|
| 948 |
-
"completion_length":
|
| 949 |
"epoch": 0.09733333333333333,
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"kl": 0.
|
| 952 |
"learning_rate": 8.467203366908707e-08,
|
| 953 |
"loss": 0.0001,
|
| 954 |
-
"reward": 1.
|
| 955 |
-
"reward_std": 0.
|
| 956 |
-
"rewards/emotion_reward_func": 0.
|
| 957 |
-
"rewards/format_reward_func":
|
| 958 |
"step": 73
|
| 959 |
},
|
| 960 |
{
|
| 961 |
-
"completion_length":
|
| 962 |
"epoch": 0.09866666666666667,
|
| 963 |
-
"grad_norm":
|
| 964 |
-
"kl": 0.
|
| 965 |
"learning_rate": 7.886322351782782e-08,
|
| 966 |
-
"loss": 0.
|
| 967 |
-
"reward":
|
| 968 |
-
"reward_std": 0.
|
| 969 |
-
"rewards/emotion_reward_func": 0.
|
| 970 |
-
"rewards/format_reward_func": 0.
|
| 971 |
"step": 74
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"completion_length":
|
| 975 |
"epoch": 0.1,
|
| 976 |
-
"grad_norm":
|
| 977 |
-
"kl": 0.
|
| 978 |
"learning_rate": 7.322330470336313e-08,
|
| 979 |
"loss": 0.0001,
|
| 980 |
-
"reward": 1.
|
| 981 |
-
"reward_std": 0.
|
| 982 |
-
"rewards/emotion_reward_func": 0.
|
| 983 |
-
"rewards/format_reward_func":
|
| 984 |
"step": 75
|
| 985 |
},
|
| 986 |
{
|
| 987 |
-
"completion_length":
|
| 988 |
"epoch": 0.10133333333333333,
|
| 989 |
-
"grad_norm":
|
| 990 |
-
"kl": 0.
|
| 991 |
"learning_rate": 6.775784314464716e-08,
|
| 992 |
"loss": 0.0001,
|
| 993 |
-
"reward":
|
| 994 |
-
"reward_std": 0.
|
| 995 |
-
"rewards/emotion_reward_func": 0.
|
| 996 |
-
"rewards/format_reward_func": 0
|
| 997 |
"step": 76
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
-
"completion_length":
|
| 1001 |
"epoch": 0.10266666666666667,
|
| 1002 |
-
"grad_norm":
|
| 1003 |
-
"kl": 0.
|
| 1004 |
"learning_rate": 6.24722325923851e-08,
|
| 1005 |
"loss": 0.0002,
|
| 1006 |
-
"reward": 1.
|
| 1007 |
-
"reward_std": 0.
|
| 1008 |
-
"rewards/emotion_reward_func": 0.
|
| 1009 |
"rewards/format_reward_func": 1.0,
|
| 1010 |
"step": 77
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
-
"completion_length":
|
| 1014 |
"epoch": 0.104,
|
| 1015 |
-
"grad_norm":
|
| 1016 |
-
"kl": 0.
|
| 1017 |
"learning_rate": 5.737168930605271e-08,
|
| 1018 |
"loss": 0.0002,
|
| 1019 |
-
"reward":
|
| 1020 |
-
"reward_std": 0.
|
| 1021 |
-
"rewards/emotion_reward_func": 0.
|
| 1022 |
"rewards/format_reward_func": 0.96875,
|
| 1023 |
"step": 78
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"completion_length":
|
| 1027 |
"epoch": 0.10533333333333333,
|
| 1028 |
-
"grad_norm":
|
| 1029 |
-
"kl": 0.
|
| 1030 |
"learning_rate": 5.246124690607739e-08,
|
| 1031 |
-
"loss": 0.
|
| 1032 |
-
"reward":
|
| 1033 |
-
"reward_std": 0.
|
| 1034 |
-
"rewards/emotion_reward_func": 0.
|
| 1035 |
-
"rewards/format_reward_func": 0
|
| 1036 |
"step": 79
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
-
"completion_length":
|
| 1040 |
"epoch": 0.10666666666666667,
|
| 1041 |
-
"grad_norm":
|
| 1042 |
-
"kl": 0.
|
| 1043 |
"learning_rate": 4.774575140626316e-08,
|
| 1044 |
-
"loss": 0.
|
| 1045 |
-
"reward":
|
| 1046 |
-
"reward_std": 0.
|
| 1047 |
-
"rewards/emotion_reward_func": 0.
|
| 1048 |
-
"rewards/format_reward_func": 0
|
| 1049 |
"step": 80
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
-
"completion_length":
|
| 1053 |
"epoch": 0.108,
|
| 1054 |
-
"grad_norm":
|
| 1055 |
-
"kl": 0.
|
| 1056 |
"learning_rate": 4.3229856431359513e-08,
|
| 1057 |
-
"loss": 0.
|
| 1058 |
-
"reward": 1.
|
| 1059 |
-
"reward_std": 0.
|
| 1060 |
-
"rewards/emotion_reward_func": 0.
|
| 1061 |
-
"rewards/format_reward_func":
|
| 1062 |
"step": 81
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
-
"completion_length":
|
| 1066 |
"epoch": 0.10933333333333334,
|
| 1067 |
-
"grad_norm":
|
| 1068 |
-
"kl": 0.
|
| 1069 |
"learning_rate": 3.8918018624496286e-08,
|
| 1070 |
"loss": 0.0001,
|
| 1071 |
-
"reward":
|
| 1072 |
-
"reward_std": 0.
|
| 1073 |
-
"rewards/emotion_reward_func": 0.
|
| 1074 |
-
"rewards/format_reward_func": 0.
|
| 1075 |
"step": 82
|
| 1076 |
},
|
| 1077 |
{
|
| 1078 |
-
"completion_length":
|
| 1079 |
"epoch": 0.11066666666666666,
|
| 1080 |
-
"grad_norm":
|
| 1081 |
-
"kl": 0.
|
| 1082 |
"learning_rate": 3.481449324901411e-08,
|
| 1083 |
"loss": 0.0001,
|
| 1084 |
-
"reward":
|
| 1085 |
-
"reward_std": 0.
|
| 1086 |
-
"rewards/emotion_reward_func": 0.
|
| 1087 |
"rewards/format_reward_func": 0.96875,
|
| 1088 |
"step": 83
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
-
"completion_length":
|
| 1092 |
"epoch": 0.112,
|
| 1093 |
-
"grad_norm":
|
| 1094 |
-
"kl": 0.
|
| 1095 |
"learning_rate": 3.092332998903416e-08,
|
| 1096 |
-
"loss": 0.
|
| 1097 |
-
"reward": 1.
|
| 1098 |
-
"reward_std": 0.
|
| 1099 |
-
"rewards/emotion_reward_func": 0.
|
| 1100 |
-
"rewards/format_reward_func":
|
| 1101 |
"step": 84
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
-
"completion_length":
|
| 1105 |
"epoch": 0.11333333333333333,
|
| 1106 |
-
"grad_norm":
|
| 1107 |
-
"kl": 0.
|
| 1108 |
"learning_rate": 2.724836895290805e-08,
|
| 1109 |
"loss": 0.0001,
|
| 1110 |
-
"reward": 1.
|
| 1111 |
-
"reward_std": 0.
|
| 1112 |
-
"rewards/emotion_reward_func": 0.
|
| 1113 |
"rewards/format_reward_func": 1.0,
|
| 1114 |
"step": 85
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
-
"completion_length":
|
| 1118 |
"epoch": 0.11466666666666667,
|
| 1119 |
-
"grad_norm":
|
| 1120 |
-
"kl": 0.
|
| 1121 |
"learning_rate": 2.379323688349516e-08,
|
| 1122 |
"loss": 0.0001,
|
| 1123 |
-
"reward": 1.
|
| 1124 |
-
"reward_std": 0.
|
| 1125 |
-
"rewards/emotion_reward_func": 0.
|
| 1126 |
"rewards/format_reward_func": 1.0,
|
| 1127 |
"step": 86
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
-
"completion_length":
|
| 1131 |
"epoch": 0.116,
|
| 1132 |
-
"grad_norm":
|
| 1133 |
-
"kl": 0.
|
| 1134 |
"learning_rate": 2.0561343579004715e-08,
|
| 1135 |
"loss": 0.0001,
|
| 1136 |
-
"reward":
|
| 1137 |
-
"reward_std": 0.
|
| 1138 |
-
"rewards/emotion_reward_func": 0.
|
| 1139 |
-
"rewards/format_reward_func": 0.
|
| 1140 |
"step": 87
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
-
"completion_length":
|
| 1144 |
"epoch": 0.11733333333333333,
|
| 1145 |
-
"grad_norm":
|
| 1146 |
-
"kl": 0.
|
| 1147 |
"learning_rate": 1.7555878527937163e-08,
|
| 1148 |
-
"loss": 0.
|
| 1149 |
-
"reward":
|
| 1150 |
-
"reward_std": 0.
|
| 1151 |
-
"rewards/emotion_reward_func": 0.
|
| 1152 |
-
"rewards/format_reward_func": 0.
|
| 1153 |
"step": 88
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
-
"completion_length":
|
| 1157 |
"epoch": 0.11866666666666667,
|
| 1158 |
-
"grad_norm":
|
| 1159 |
-
"kl": 0.
|
| 1160 |
"learning_rate": 1.4779807761443635e-08,
|
| 1161 |
"loss": 0.0001,
|
| 1162 |
-
"reward":
|
| 1163 |
-
"reward_std": 0.
|
| 1164 |
-
"rewards/emotion_reward_func": 0.
|
| 1165 |
-
"rewards/format_reward_func": 0.
|
| 1166 |
"step": 89
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
-
"completion_length":
|
| 1170 |
"epoch": 0.12,
|
| 1171 |
-
"grad_norm": 7.
|
| 1172 |
-
"kl": 0.
|
| 1173 |
"learning_rate": 1.2235870926211616e-08,
|
| 1174 |
"loss": 0.0001,
|
| 1175 |
-
"reward":
|
| 1176 |
-
"reward_std": 0.
|
| 1177 |
-
"rewards/emotion_reward_func": 0.
|
| 1178 |
-
"rewards/format_reward_func": 0
|
| 1179 |
"step": 90
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
-
"completion_length":
|
| 1183 |
"epoch": 0.12133333333333333,
|
| 1184 |
-
"grad_norm":
|
| 1185 |
-
"kl": 0.
|
| 1186 |
"learning_rate": 9.926578580764234e-09,
|
| 1187 |
"loss": 0.0001,
|
| 1188 |
-
"reward":
|
| 1189 |
-
"reward_std": 0.
|
| 1190 |
-
"rewards/emotion_reward_func": 0.
|
| 1191 |
-
"rewards/format_reward_func": 0
|
| 1192 |
"step": 91
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
-
"completion_length":
|
| 1196 |
"epoch": 0.12266666666666666,
|
| 1197 |
-
"grad_norm":
|
| 1198 |
-
"kl": 0.
|
| 1199 |
"learning_rate": 7.85420971784223e-09,
|
| 1200 |
"loss": 0.0001,
|
| 1201 |
-
"reward":
|
| 1202 |
-
"reward_std": 0.
|
| 1203 |
-
"rewards/emotion_reward_func": 0.
|
| 1204 |
-
"rewards/format_reward_func": 0
|
| 1205 |
"step": 92
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
-
"completion_length":
|
| 1209 |
"epoch": 0.124,
|
| 1210 |
-
"grad_norm":
|
| 1211 |
-
"kl": 0.
|
| 1212 |
"learning_rate": 6.020809515313141e-09,
|
| 1213 |
"loss": 0.0001,
|
| 1214 |
-
"reward":
|
| 1215 |
-
"reward_std": 0.
|
| 1216 |
-
"rewards/emotion_reward_func": 0.
|
| 1217 |
-
"rewards/format_reward_func": 0
|
| 1218 |
"step": 93
|
| 1219 |
},
|
| 1220 |
{
|
| 1221 |
-
"completion_length":
|
| 1222 |
"epoch": 0.12533333333333332,
|
| 1223 |
-
"grad_norm":
|
| 1224 |
-
"kl": 0.
|
| 1225 |
"learning_rate": 4.4281873178278475e-09,
|
| 1226 |
"loss": 0.0002,
|
| 1227 |
-
"reward":
|
| 1228 |
-
"reward_std": 0.
|
| 1229 |
-
"rewards/emotion_reward_func": 0.
|
| 1230 |
-
"rewards/format_reward_func": 0.
|
| 1231 |
"step": 94
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
-
"completion_length":
|
| 1235 |
"epoch": 0.12666666666666668,
|
| 1236 |
-
"grad_norm":
|
| 1237 |
-
"kl": 0.
|
| 1238 |
"learning_rate": 3.077914851215585e-09,
|
| 1239 |
-
"loss": 0.
|
| 1240 |
-
"reward":
|
| 1241 |
-
"reward_std": 0.
|
| 1242 |
-
"rewards/emotion_reward_func": 0.
|
| 1243 |
-
"rewards/format_reward_func": 0
|
| 1244 |
"step": 95
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
-
"completion_length":
|
| 1248 |
"epoch": 0.128,
|
| 1249 |
-
"grad_norm":
|
| 1250 |
-
"kl": 0.
|
| 1251 |
"learning_rate": 1.9713246713805587e-09,
|
| 1252 |
"loss": 0.0001,
|
| 1253 |
-
"reward":
|
| 1254 |
-
"reward_std": 0.
|
| 1255 |
-
"rewards/emotion_reward_func": 0.
|
| 1256 |
-
"rewards/format_reward_func": 0.
|
| 1257 |
"step": 96
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
-
"completion_length":
|
| 1261 |
"epoch": 0.12933333333333333,
|
| 1262 |
-
"grad_norm":
|
| 1263 |
-
"kl": 0.
|
| 1264 |
"learning_rate": 1.1095088492300008e-09,
|
| 1265 |
-
"loss": 0.
|
| 1266 |
-
"reward":
|
| 1267 |
-
"reward_std": 0.
|
| 1268 |
-
"rewards/emotion_reward_func": 0.
|
| 1269 |
-
"rewards/format_reward_func": 0
|
| 1270 |
"step": 97
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
-
"completion_length":
|
| 1274 |
"epoch": 0.13066666666666665,
|
| 1275 |
-
"grad_norm":
|
| 1276 |
-
"kl": 0.
|
| 1277 |
"learning_rate": 4.933178929321102e-10,
|
| 1278 |
-
"loss": 0.
|
| 1279 |
-
"reward": 1.
|
| 1280 |
-
"reward_std": 0.
|
| 1281 |
-
"rewards/emotion_reward_func": 0.
|
| 1282 |
-
"rewards/format_reward_func":
|
| 1283 |
"step": 98
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
-
"completion_length":
|
| 1287 |
"epoch": 0.132,
|
| 1288 |
-
"grad_norm":
|
| 1289 |
-
"kl": 0.
|
| 1290 |
"learning_rate": 1.2335990856709998e-10,
|
| 1291 |
"loss": 0.0001,
|
| 1292 |
-
"reward": 1.
|
| 1293 |
-
"reward_std": 0.
|
| 1294 |
-
"rewards/emotion_reward_func": 0.
|
| 1295 |
-
"rewards/format_reward_func":
|
| 1296 |
"step": 99
|
| 1297 |
},
|
| 1298 |
{
|
| 1299 |
-
"completion_length":
|
| 1300 |
"epoch": 0.13333333333333333,
|
| 1301 |
-
"grad_norm":
|
| 1302 |
-
"kl": 0.
|
| 1303 |
"learning_rate": 0.0,
|
| 1304 |
"loss": 0.0001,
|
| 1305 |
-
"reward": 1.
|
| 1306 |
-
"reward_std": 0.
|
| 1307 |
-
"rewards/emotion_reward_func": 0.
|
| 1308 |
-
"rewards/format_reward_func":
|
| 1309 |
"step": 100
|
| 1310 |
}
|
| 1311 |
],
|
|
@@ -1327,7 +1327,7 @@
|
|
| 1327 |
}
|
| 1328 |
},
|
| 1329 |
"total_flos": 0.0,
|
| 1330 |
-
"train_batch_size":
|
| 1331 |
"trial_name": null,
|
| 1332 |
"trial_params": null
|
| 1333 |
}
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"completion_length": 113.171875,
|
| 13 |
"epoch": 0.0013333333333333333,
|
| 14 |
"grad_norm": 0.0,
|
| 15 |
"kl": 0.0,
|
|
|
|
| 22 |
"step": 1
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"completion_length": 100.609375,
|
| 26 |
"epoch": 0.0026666666666666666,
|
| 27 |
+
"grad_norm": 7.206315517425537,
|
| 28 |
"kl": 0.0,
|
| 29 |
"learning_rate": 4.995066821070679e-07,
|
| 30 |
+
"loss": 0.0,
|
| 31 |
+
"reward": 0.046875,
|
| 32 |
+
"reward_std": 0.1875,
|
| 33 |
+
"rewards/emotion_reward_func": 0.015625,
|
| 34 |
+
"rewards/format_reward_func": 0.03125,
|
| 35 |
"step": 2
|
| 36 |
},
|
| 37 |
{
|
| 38 |
+
"completion_length": 130.65625,
|
| 39 |
"epoch": 0.004,
|
| 40 |
+
"grad_norm": 7.190981864929199,
|
| 41 |
+
"kl": 0.0008925930160330608,
|
| 42 |
"learning_rate": 4.9889049115077e-07,
|
| 43 |
"loss": 0.0,
|
| 44 |
+
"reward": 0.0625,
|
| 45 |
+
"reward_std": 0.25,
|
| 46 |
+
"rewards/emotion_reward_func": 0.015625,
|
| 47 |
+
"rewards/format_reward_func": 0.046875,
|
| 48 |
"step": 3
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"completion_length": 112.484375,
|
| 52 |
"epoch": 0.005333333333333333,
|
| 53 |
+
"grad_norm": 4.745754718780518,
|
| 54 |
+
"kl": 0.0009372199856443331,
|
| 55 |
"learning_rate": 4.980286753286194e-07,
|
| 56 |
"loss": 0.0,
|
| 57 |
+
"reward": 0.046875,
|
| 58 |
+
"reward_std": 0.1875,
|
| 59 |
+
"rewards/emotion_reward_func": 0.015625,
|
| 60 |
+
"rewards/format_reward_func": 0.03125,
|
| 61 |
"step": 4
|
| 62 |
},
|
| 63 |
{
|
| 64 |
+
"completion_length": 100.359375,
|
| 65 |
"epoch": 0.006666666666666667,
|
| 66 |
+
"grad_norm": 6.957683086395264,
|
| 67 |
+
"kl": 0.0009052860696101561,
|
| 68 |
"learning_rate": 4.969220851487844e-07,
|
| 69 |
"loss": 0.0,
|
| 70 |
+
"reward": 0.078125,
|
| 71 |
+
"reward_std": 0.3125,
|
| 72 |
+
"rewards/emotion_reward_func": 0.03125,
|
| 73 |
+
"rewards/format_reward_func": 0.046875,
|
| 74 |
"step": 5
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"completion_length": 109.5625,
|
| 78 |
"epoch": 0.008,
|
| 79 |
+
"grad_norm": 8.517982482910156,
|
| 80 |
+
"kl": 0.0011545967281563208,
|
| 81 |
"learning_rate": 4.955718126821722e-07,
|
| 82 |
"loss": 0.0,
|
| 83 |
+
"reward": 0.140625,
|
| 84 |
+
"reward_std": 0.34228479862213135,
|
| 85 |
+
"rewards/emotion_reward_func": 0.03125,
|
| 86 |
+
"rewards/format_reward_func": 0.109375,
|
| 87 |
"step": 6
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"completion_length": 117.484375,
|
| 91 |
"epoch": 0.009333333333333334,
|
| 92 |
+
"grad_norm": 2.9639530181884766,
|
| 93 |
+
"kl": 0.001261903322301805,
|
| 94 |
"learning_rate": 4.939791904846868e-07,
|
| 95 |
"loss": 0.0,
|
| 96 |
+
"reward": 0.03125,
|
| 97 |
"reward_std": 0.125,
|
| 98 |
+
"rewards/emotion_reward_func": 0.015625,
|
| 99 |
+
"rewards/format_reward_func": 0.015625,
|
| 100 |
"step": 7
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"completion_length": 118.234375,
|
| 104 |
"epoch": 0.010666666666666666,
|
| 105 |
+
"grad_norm": 8.49864387512207,
|
| 106 |
+
"kl": 0.002180565701564774,
|
| 107 |
"learning_rate": 4.921457902821578e-07,
|
| 108 |
"loss": 0.0,
|
| 109 |
+
"reward": 0.140625,
|
| 110 |
+
"reward_std": 0.4436737596988678,
|
| 111 |
+
"rewards/emotion_reward_func": 0.046875,
|
| 112 |
+
"rewards/format_reward_func": 0.09375,
|
| 113 |
"step": 8
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"completion_length": 106.546875,
|
| 117 |
"epoch": 0.012,
|
| 118 |
+
"grad_norm": 7.236983299255371,
|
| 119 |
+
"kl": 0.0018676594190765172,
|
| 120 |
"learning_rate": 4.900734214192358e-07,
|
| 121 |
"loss": 0.0,
|
| 122 |
+
"reward": 0.109375,
|
| 123 |
+
"reward_std": 0.3186737596988678,
|
| 124 |
+
"rewards/emotion_reward_func": 0.03125,
|
| 125 |
+
"rewards/format_reward_func": 0.078125,
|
| 126 |
"step": 9
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"completion_length": 94.640625,
|
| 130 |
"epoch": 0.013333333333333334,
|
| 131 |
+
"grad_norm": 6.612936973571777,
|
| 132 |
+
"kl": 0.00287746504181996,
|
| 133 |
"learning_rate": 4.877641290737883e-07,
|
| 134 |
"loss": 0.0,
|
| 135 |
+
"reward": 0.15625,
|
| 136 |
+
"reward_std": 0.38375620543956757,
|
| 137 |
+
"rewards/emotion_reward_func": 0.03125,
|
| 138 |
+
"rewards/format_reward_func": 0.125,
|
| 139 |
"step": 10
|
| 140 |
},
|
| 141 |
{
|
| 142 |
+
"completion_length": 78.453125,
|
| 143 |
"epoch": 0.014666666666666666,
|
| 144 |
+
"grad_norm": 6.714486598968506,
|
| 145 |
+
"kl": 0.0037596136680804193,
|
| 146 |
"learning_rate": 4.852201922385564e-07,
|
| 147 |
"loss": 0.0,
|
| 148 |
+
"reward": 0.15625,
|
| 149 |
+
"reward_std": 0.34297704696655273,
|
| 150 |
+
"rewards/emotion_reward_func": 0.03125,
|
| 151 |
+
"rewards/format_reward_func": 0.125,
|
| 152 |
"step": 11
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"completion_length": 113.765625,
|
| 156 |
"epoch": 0.016,
|
| 157 |
+
"grad_norm": 6.222474098205566,
|
| 158 |
+
"kl": 0.005189417512156069,
|
| 159 |
"learning_rate": 4.824441214720628e-07,
|
| 160 |
"loss": 0.0,
|
| 161 |
+
"reward": 0.328125,
|
| 162 |
+
"reward_std": 0.6560364216566086,
|
| 163 |
+
"rewards/emotion_reward_func": 0.109375,
|
| 164 |
+
"rewards/format_reward_func": 0.21875,
|
| 165 |
"step": 12
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"completion_length": 93.890625,
|
| 169 |
"epoch": 0.017333333333333333,
|
| 170 |
+
"grad_norm": 7.199591636657715,
|
| 171 |
+
"kl": 0.011019598576240242,
|
| 172 |
"learning_rate": 4.794386564209952e-07,
|
| 173 |
"loss": 0.0,
|
| 174 |
+
"reward": 0.359375,
|
| 175 |
+
"reward_std": 0.647876650094986,
|
| 176 |
+
"rewards/emotion_reward_func": 0.125,
|
| 177 |
+
"rewards/format_reward_func": 0.234375,
|
| 178 |
"step": 13
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"completion_length": 93.6875,
|
| 182 |
"epoch": 0.018666666666666668,
|
| 183 |
+
"grad_norm": 7.459893226623535,
|
| 184 |
+
"kl": 0.012794358422979712,
|
| 185 |
"learning_rate": 4.762067631165049e-07,
|
| 186 |
"loss": 0.0,
|
| 187 |
+
"reward": 0.28125,
|
| 188 |
+
"reward_std": 0.4561576098203659,
|
| 189 |
+
"rewards/emotion_reward_func": 0.015625,
|
| 190 |
+
"rewards/format_reward_func": 0.265625,
|
| 191 |
"step": 14
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"completion_length": 114.03125,
|
| 195 |
"epoch": 0.02,
|
| 196 |
+
"grad_norm": 7.204787731170654,
|
| 197 |
+
"kl": 0.01207199739292264,
|
| 198 |
"learning_rate": 4.7275163104709194e-07,
|
| 199 |
+
"loss": 0.0,
|
| 200 |
+
"reward": 0.421875,
|
| 201 |
+
"reward_std": 0.6076867878437042,
|
| 202 |
+
"rewards/emotion_reward_func": 0.0625,
|
| 203 |
+
"rewards/format_reward_func": 0.359375,
|
| 204 |
"step": 15
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"completion_length": 90.1875,
|
| 208 |
"epoch": 0.021333333333333333,
|
| 209 |
+
"grad_norm": 6.9398369789123535,
|
| 210 |
+
"kl": 0.012930819648317993,
|
| 211 |
"learning_rate": 4.6907667001096585e-07,
|
| 212 |
"loss": 0.0,
|
| 213 |
+
"reward": 0.296875,
|
| 214 |
+
"reward_std": 0.6791985481977463,
|
| 215 |
+
"rewards/emotion_reward_func": 0.125,
|
| 216 |
+
"rewards/format_reward_func": 0.171875,
|
| 217 |
"step": 16
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"completion_length": 100.265625,
|
| 221 |
"epoch": 0.02266666666666667,
|
| 222 |
+
"grad_norm": 7.01741361618042,
|
| 223 |
+
"kl": 0.014807499013841152,
|
| 224 |
"learning_rate": 4.6518550675098587e-07,
|
| 225 |
"loss": 0.0,
|
| 226 |
+
"reward": 0.609375,
|
| 227 |
+
"reward_std": 0.7960269749164581,
|
| 228 |
+
"rewards/emotion_reward_func": 0.1875,
|
| 229 |
+
"rewards/format_reward_func": 0.421875,
|
| 230 |
"step": 17
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"completion_length": 101.59375,
|
| 234 |
"epoch": 0.024,
|
| 235 |
+
"grad_norm": 6.980762004852295,
|
| 236 |
+
"kl": 0.02019192511215806,
|
| 237 |
"learning_rate": 4.6108198137550377e-07,
|
| 238 |
"loss": 0.0,
|
| 239 |
+
"reward": 0.53125,
|
| 240 |
+
"reward_std": 0.6885540634393692,
|
| 241 |
+
"rewards/emotion_reward_func": 0.125,
|
| 242 |
+
"rewards/format_reward_func": 0.40625,
|
| 243 |
"step": 18
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"completion_length": 93.859375,
|
| 247 |
"epoch": 0.025333333333333333,
|
| 248 |
+
"grad_norm": 8.390938758850098,
|
| 249 |
+
"kl": 0.03694334626197815,
|
| 250 |
"learning_rate": 4.567701435686404e-07,
|
| 251 |
"loss": 0.0,
|
| 252 |
+
"reward": 0.65625,
|
| 253 |
+
"reward_std": 0.651764303445816,
|
| 254 |
+
"rewards/emotion_reward_func": 0.125,
|
| 255 |
+
"rewards/format_reward_func": 0.53125,
|
| 256 |
"step": 19
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"completion_length": 85.546875,
|
| 260 |
"epoch": 0.02666666666666667,
|
| 261 |
+
"grad_norm": 8.39484691619873,
|
| 262 |
+
"kl": 0.03380461875349283,
|
| 263 |
"learning_rate": 4.5225424859373684e-07,
|
| 264 |
"loss": 0.0,
|
| 265 |
+
"reward": 0.765625,
|
| 266 |
+
"reward_std": 0.7304560542106628,
|
| 267 |
+
"rewards/emotion_reward_func": 0.203125,
|
| 268 |
+
"rewards/format_reward_func": 0.5625,
|
| 269 |
"step": 20
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"completion_length": 91.3125,
|
| 273 |
"epoch": 0.028,
|
| 274 |
+
"grad_norm": 6.440164089202881,
|
| 275 |
+
"kl": 0.04492605570703745,
|
| 276 |
"learning_rate": 4.475387530939226e-07,
|
| 277 |
"loss": 0.0,
|
| 278 |
+
"reward": 0.765625,
|
| 279 |
+
"reward_std": 0.5731314420700073,
|
| 280 |
+
"rewards/emotion_reward_func": 0.078125,
|
| 281 |
+
"rewards/format_reward_func": 0.6875,
|
| 282 |
"step": 21
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"completion_length": 91.0625,
|
| 286 |
"epoch": 0.029333333333333333,
|
| 287 |
+
"grad_norm": 6.099334716796875,
|
| 288 |
+
"kl": 0.040256964042782784,
|
| 289 |
"learning_rate": 4.426283106939473e-07,
|
| 290 |
"loss": 0.0,
|
| 291 |
+
"reward": 1.1875,
|
| 292 |
+
"reward_std": 0.6744011342525482,
|
| 293 |
+
"rewards/emotion_reward_func": 0.390625,
|
| 294 |
+
"rewards/format_reward_func": 0.796875,
|
| 295 |
"step": 22
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"completion_length": 79.328125,
|
| 299 |
"epoch": 0.030666666666666665,
|
| 300 |
+
"grad_norm": 7.1284871101379395,
|
| 301 |
+
"kl": 0.04640346672385931,
|
| 302 |
"learning_rate": 4.375277674076149e-07,
|
| 303 |
"loss": 0.0,
|
| 304 |
+
"reward": 1.21875,
|
| 305 |
+
"reward_std": 0.7463032901287079,
|
| 306 |
+
"rewards/emotion_reward_func": 0.4375,
|
| 307 |
+
"rewards/format_reward_func": 0.78125,
|
| 308 |
"step": 23
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"completion_length": 71.84375,
|
| 312 |
"epoch": 0.032,
|
| 313 |
+
"grad_norm": 6.066468715667725,
|
| 314 |
+
"kl": 0.07137730903923512,
|
| 315 |
"learning_rate": 4.3224215685535287e-07,
|
| 316 |
+
"loss": 0.0001,
|
| 317 |
+
"reward": 1.328125,
|
| 318 |
+
"reward_std": 0.8113406747579575,
|
| 319 |
+
"rewards/emotion_reward_func": 0.546875,
|
| 320 |
+
"rewards/format_reward_func": 0.78125,
|
| 321 |
"step": 24
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"completion_length": 101.0625,
|
| 325 |
"epoch": 0.03333333333333333,
|
| 326 |
+
"grad_norm": 5.047848224639893,
|
| 327 |
+
"kl": 0.0665823919698596,
|
| 328 |
"learning_rate": 4.2677669529663686e-07,
|
| 329 |
+
"loss": 0.0001,
|
| 330 |
+
"reward": 0.828125,
|
| 331 |
+
"reward_std": 0.5247472077608109,
|
| 332 |
+
"rewards/emotion_reward_func": 0.078125,
|
| 333 |
+
"rewards/format_reward_func": 0.75,
|
| 334 |
"step": 25
|
| 335 |
},
|
| 336 |
{
|
| 337 |
+
"completion_length": 68.1875,
|
| 338 |
"epoch": 0.034666666666666665,
|
| 339 |
+
"grad_norm": 6.7862958908081055,
|
| 340 |
+
"kl": 0.07357331551611423,
|
| 341 |
"learning_rate": 4.2113677648217216e-07,
|
| 342 |
"loss": 0.0001,
|
| 343 |
+
"reward": 1.515625,
|
| 344 |
+
"reward_std": 0.7178780436515808,
|
| 345 |
+
"rewards/emotion_reward_func": 0.671875,
|
| 346 |
+
"rewards/format_reward_func": 0.84375,
|
| 347 |
"step": 26
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"completion_length": 73.5625,
|
| 351 |
"epoch": 0.036,
|
| 352 |
+
"grad_norm": 6.041502475738525,
|
| 353 |
+
"kl": 0.0839837146922946,
|
| 354 |
"learning_rate": 4.1532796633091294e-07,
|
| 355 |
+
"loss": 0.0001,
|
| 356 |
+
"reward": 1.0,
|
| 357 |
+
"reward_std": 0.5787727609276772,
|
| 358 |
+
"rewards/emotion_reward_func": 0.1875,
|
| 359 |
+
"rewards/format_reward_func": 0.8125,
|
| 360 |
"step": 27
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"completion_length": 75.515625,
|
| 364 |
"epoch": 0.037333333333333336,
|
| 365 |
+
"grad_norm": 5.880768299102783,
|
| 366 |
+
"kl": 0.07887133583426476,
|
| 367 |
"learning_rate": 4.0935599743717244e-07,
|
| 368 |
"loss": 0.0001,
|
| 369 |
+
"reward": 1.265625,
|
| 370 |
+
"reward_std": 0.6387931928038597,
|
| 371 |
+
"rewards/emotion_reward_func": 0.390625,
|
| 372 |
+
"rewards/format_reward_func": 0.875,
|
| 373 |
"step": 28
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"completion_length": 84.65625,
|
| 377 |
"epoch": 0.03866666666666667,
|
| 378 |
+
"grad_norm": 4.456689357757568,
|
| 379 |
+
"kl": 0.05541729833930731,
|
| 380 |
"learning_rate": 4.0322676341324414e-07,
|
| 381 |
+
"loss": 0.0001,
|
| 382 |
+
"reward": 1.546875,
|
| 383 |
+
"reward_std": 0.6021395623683929,
|
| 384 |
+
"rewards/emotion_reward_func": 0.6875,
|
| 385 |
+
"rewards/format_reward_func": 0.859375,
|
| 386 |
"step": 29
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"completion_length": 72.03125,
|
| 390 |
"epoch": 0.04,
|
| 391 |
+
"grad_norm": 6.281215667724609,
|
| 392 |
+
"kl": 0.07642886973917484,
|
| 393 |
"learning_rate": 3.9694631307311825e-07,
|
| 394 |
+
"loss": 0.0001,
|
| 395 |
+
"reward": 1.421875,
|
| 396 |
+
"reward_std": 0.6625982969999313,
|
| 397 |
+
"rewards/emotion_reward_func": 0.515625,
|
| 398 |
+
"rewards/format_reward_func": 0.90625,
|
| 399 |
"step": 30
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"completion_length": 64.96875,
|
| 403 |
"epoch": 0.04133333333333333,
|
| 404 |
+
"grad_norm": 6.267475128173828,
|
| 405 |
+
"kl": 0.07545926049351692,
|
| 406 |
"learning_rate": 3.9052084446303265e-07,
|
| 407 |
"loss": 0.0001,
|
| 408 |
+
"reward": 1.28125,
|
| 409 |
+
"reward_std": 0.5209204778075218,
|
| 410 |
+
"rewards/emotion_reward_func": 0.34375,
|
| 411 |
+
"rewards/format_reward_func": 0.9375,
|
| 412 |
"step": 31
|
| 413 |
},
|
| 414 |
{
|
| 415 |
+
"completion_length": 74.28125,
|
| 416 |
"epoch": 0.042666666666666665,
|
| 417 |
+
"grad_norm": 6.10364294052124,
|
| 418 |
+
"kl": 0.08546704892069101,
|
| 419 |
"learning_rate": 3.839566987447491e-07,
|
| 420 |
+
"loss": 0.0001,
|
| 421 |
+
"reward": 1.59375,
|
| 422 |
+
"reward_std": 0.7168828397989273,
|
| 423 |
+
"rewards/emotion_reward_func": 0.71875,
|
| 424 |
+
"rewards/format_reward_func": 0.875,
|
| 425 |
"step": 32
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"completion_length": 70.03125,
|
| 429 |
"epoch": 0.044,
|
| 430 |
+
"grad_norm": 4.039772033691406,
|
| 431 |
+
"kl": 0.06002845522016287,
|
| 432 |
"learning_rate": 3.7726035393759283e-07,
|
| 433 |
"loss": 0.0001,
|
| 434 |
+
"reward": 1.3125,
|
| 435 |
+
"reward_std": 0.2808031141757965,
|
| 436 |
+
"rewards/emotion_reward_func": 0.390625,
|
| 437 |
+
"rewards/format_reward_func": 0.921875,
|
| 438 |
"step": 33
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"completion_length": 83.109375,
|
| 442 |
"epoch": 0.04533333333333334,
|
| 443 |
+
"grad_norm": 6.683216094970703,
|
| 444 |
+
"kl": 0.0706396009773016,
|
| 445 |
"learning_rate": 3.704384185254288e-07,
|
| 446 |
"loss": 0.0001,
|
| 447 |
+
"reward": 1.421875,
|
| 448 |
+
"reward_std": 0.7211004346609116,
|
| 449 |
+
"rewards/emotion_reward_func": 0.578125,
|
| 450 |
+
"rewards/format_reward_func": 0.84375,
|
| 451 |
"step": 34
|
| 452 |
},
|
| 453 |
{
|
| 454 |
+
"completion_length": 72.765625,
|
| 455 |
"epoch": 0.04666666666666667,
|
| 456 |
+
"grad_norm": 6.402747631072998,
|
| 457 |
+
"kl": 0.10811681114137173,
|
| 458 |
"learning_rate": 3.634976249348867e-07,
|
| 459 |
"loss": 0.0001,
|
| 460 |
+
"reward": 1.28125,
|
| 461 |
+
"reward_std": 0.5904398858547211,
|
| 462 |
+
"rewards/emotion_reward_func": 0.375,
|
| 463 |
+
"rewards/format_reward_func": 0.90625,
|
| 464 |
"step": 35
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"completion_length": 78.734375,
|
| 468 |
"epoch": 0.048,
|
| 469 |
+
"grad_norm": 5.202386856079102,
|
| 470 |
+
"kl": 0.07850308250635862,
|
| 471 |
"learning_rate": 3.5644482289126813e-07,
|
| 472 |
"loss": 0.0001,
|
| 473 |
+
"reward": 1.09375,
|
| 474 |
+
"reward_std": 0.5113069340586662,
|
| 475 |
+
"rewards/emotion_reward_func": 0.203125,
|
| 476 |
+
"rewards/format_reward_func": 0.890625,
|
| 477 |
"step": 36
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"completion_length": 68.953125,
|
| 481 |
"epoch": 0.04933333333333333,
|
| 482 |
+
"grad_norm": 7.1071085929870605,
|
| 483 |
+
"kl": 0.08383779786527157,
|
| 484 |
"learning_rate": 3.492869726586951e-07,
|
| 485 |
"loss": 0.0001,
|
| 486 |
+
"reward": 1.1875,
|
| 487 |
+
"reward_std": 0.4691474586725235,
|
| 488 |
+
"rewards/emotion_reward_func": 0.21875,
|
| 489 |
+
"rewards/format_reward_func": 0.96875,
|
| 490 |
"step": 37
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"completion_length": 65.734375,
|
| 494 |
"epoch": 0.050666666666666665,
|
| 495 |
+
"grad_norm": 6.272955417633057,
|
| 496 |
+
"kl": 0.06441066134721041,
|
| 497 |
"learning_rate": 3.4203113817116953e-07,
|
| 498 |
"loss": 0.0001,
|
| 499 |
+
"reward": 1.75,
|
| 500 |
+
"reward_std": 0.41095855832099915,
|
| 501 |
+
"rewards/emotion_reward_func": 0.765625,
|
| 502 |
+
"rewards/format_reward_func": 0.984375,
|
| 503 |
"step": 38
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"completion_length": 69.9375,
|
| 507 |
"epoch": 0.052,
|
| 508 |
+
"grad_norm": 3.7793548107147217,
|
| 509 |
+
"kl": 0.06628133170306683,
|
| 510 |
"learning_rate": 3.346844800613229e-07,
|
| 511 |
"loss": 0.0001,
|
| 512 |
+
"reward": 1.078125,
|
| 513 |
+
"reward_std": 0.21347813308238983,
|
| 514 |
+
"rewards/emotion_reward_func": 0.109375,
|
| 515 |
+
"rewards/format_reward_func": 0.96875,
|
| 516 |
"step": 39
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"completion_length": 64.34375,
|
| 520 |
"epoch": 0.05333333333333334,
|
| 521 |
+
"grad_norm": 6.500398635864258,
|
| 522 |
+
"kl": 0.06995576526969671,
|
| 523 |
"learning_rate": 3.272542485937368e-07,
|
| 524 |
"loss": 0.0001,
|
| 525 |
+
"reward": 1.546875,
|
| 526 |
+
"reward_std": 0.47020626068115234,
|
| 527 |
+
"rewards/emotion_reward_func": 0.5625,
|
| 528 |
+
"rewards/format_reward_func": 0.984375,
|
| 529 |
"step": 40
|
| 530 |
},
|
| 531 |
{
|
| 532 |
+
"completion_length": 73.375,
|
| 533 |
"epoch": 0.05466666666666667,
|
| 534 |
+
"grad_norm": 4.637302875518799,
|
| 535 |
+
"kl": 0.06445631105452776,
|
| 536 |
"learning_rate": 3.1974777650980734e-07,
|
| 537 |
"loss": 0.0001,
|
| 538 |
+
"reward": 1.578125,
|
| 539 |
+
"reward_std": 0.38923946768045425,
|
| 540 |
+
"rewards/emotion_reward_func": 0.609375,
|
| 541 |
+
"rewards/format_reward_func": 0.96875,
|
| 542 |
"step": 41
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"completion_length": 69.875,
|
| 546 |
"epoch": 0.056,
|
| 547 |
+
"grad_norm": 5.5001220703125,
|
| 548 |
+
"kl": 0.06374164298176765,
|
| 549 |
"learning_rate": 3.121724717912138e-07,
|
| 550 |
"loss": 0.0001,
|
| 551 |
+
"reward": 1.671875,
|
| 552 |
+
"reward_std": 0.18616947531700134,
|
| 553 |
+
"rewards/emotion_reward_func": 0.671875,
|
| 554 |
+
"rewards/format_reward_func": 1.0,
|
| 555 |
"step": 42
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"completion_length": 65.75,
|
| 559 |
"epoch": 0.05733333333333333,
|
| 560 |
+
"grad_norm": 5.561408996582031,
|
| 561 |
+
"kl": 0.11185399815440178,
|
| 562 |
"learning_rate": 3.0453581034913565e-07,
|
| 563 |
"loss": 0.0001,
|
| 564 |
+
"reward": 1.125,
|
| 565 |
+
"reward_std": 0.4263191595673561,
|
| 566 |
+
"rewards/emotion_reward_func": 0.1875,
|
| 567 |
+
"rewards/format_reward_func": 0.9375,
|
| 568 |
"step": 43
|
| 569 |
},
|
| 570 |
{
|
| 571 |
+
"completion_length": 57.59375,
|
| 572 |
"epoch": 0.058666666666666666,
|
| 573 |
+
"grad_norm": 6.903532028198242,
|
| 574 |
+
"kl": 0.08089348301291466,
|
| 575 |
"learning_rate": 2.968453286464312e-07,
|
| 576 |
"loss": 0.0001,
|
| 577 |
+
"reward": 1.75,
|
| 578 |
+
"reward_std": 0.3838024437427521,
|
| 579 |
+
"rewards/emotion_reward_func": 0.765625,
|
| 580 |
+
"rewards/format_reward_func": 0.984375,
|
| 581 |
"step": 44
|
| 582 |
},
|
| 583 |
{
|
| 584 |
+
"completion_length": 81.0625,
|
| 585 |
"epoch": 0.06,
|
| 586 |
+
"grad_norm": 5.9377288818359375,
|
| 587 |
+
"kl": 0.06169276125729084,
|
| 588 |
"learning_rate": 2.8910861626005773e-07,
|
| 589 |
"loss": 0.0001,
|
| 590 |
+
"reward": 1.375,
|
| 591 |
+
"reward_std": 0.5583916157484055,
|
| 592 |
+
"rewards/emotion_reward_func": 0.421875,
|
| 593 |
+
"rewards/format_reward_func": 0.953125,
|
| 594 |
"step": 45
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"completion_length": 64.09375,
|
| 598 |
"epoch": 0.06133333333333333,
|
| 599 |
+
"grad_norm": 6.421321392059326,
|
| 600 |
+
"kl": 0.08067317306995392,
|
| 601 |
"learning_rate": 2.8133330839107604e-07,
|
| 602 |
"loss": 0.0001,
|
| 603 |
+
"reward": 1.484375,
|
| 604 |
+
"reward_std": 0.6873095482587814,
|
| 605 |
+
"rewards/emotion_reward_func": 0.59375,
|
| 606 |
+
"rewards/format_reward_func": 0.890625,
|
| 607 |
"step": 46
|
| 608 |
},
|
| 609 |
{
|
| 610 |
+
"completion_length": 65.859375,
|
| 611 |
"epoch": 0.06266666666666666,
|
| 612 |
+
"grad_norm": 5.465454578399658,
|
| 613 |
+
"kl": 0.08742601610720158,
|
| 614 |
"learning_rate": 2.735270783296286e-07,
|
| 615 |
"loss": 0.0001,
|
| 616 |
+
"reward": 1.421875,
|
| 617 |
+
"reward_std": 0.37412673234939575,
|
| 618 |
+
"rewards/emotion_reward_func": 0.4375,
|
| 619 |
+
"rewards/format_reward_func": 0.984375,
|
| 620 |
"step": 47
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"completion_length": 61.734375,
|
| 624 |
"epoch": 0.064,
|
| 625 |
+
"grad_norm": 6.597202777862549,
|
| 626 |
+
"kl": 0.09020566754043102,
|
| 627 |
"learning_rate": 2.6569762988232837e-07,
|
| 628 |
"loss": 0.0001,
|
| 629 |
+
"reward": 1.203125,
|
| 630 |
+
"reward_std": 0.4003961533308029,
|
| 631 |
+
"rewards/emotion_reward_func": 0.265625,
|
| 632 |
+
"rewards/format_reward_func": 0.9375,
|
| 633 |
"step": 48
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"completion_length": 63.21875,
|
| 637 |
"epoch": 0.06533333333333333,
|
| 638 |
+
"grad_norm": 5.847234725952148,
|
| 639 |
+
"kl": 0.09095379617065191,
|
| 640 |
"learning_rate": 2.5785268976953204e-07,
|
| 641 |
"loss": 0.0001,
|
| 642 |
+
"reward": 1.375,
|
| 643 |
+
"reward_std": 0.3876233473420143,
|
| 644 |
+
"rewards/emotion_reward_func": 0.390625,
|
| 645 |
+
"rewards/format_reward_func": 0.984375,
|
| 646 |
"step": 49
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"completion_length": 59.890625,
|
| 650 |
"epoch": 0.06666666666666667,
|
| 651 |
+
"grad_norm": 7.6002936363220215,
|
| 652 |
+
"kl": 0.08561510033905506,
|
| 653 |
"learning_rate": 2.5e-07,
|
| 654 |
"loss": 0.0001,
|
| 655 |
+
"reward": 1.25,
|
| 656 |
+
"reward_std": 0.3925696462392807,
|
| 657 |
+
"rewards/emotion_reward_func": 0.28125,
|
| 658 |
+
"rewards/format_reward_func": 0.96875,
|
| 659 |
"step": 50
|
| 660 |
},
|
| 661 |
{
|
| 662 |
+
"completion_length": 57.890625,
|
| 663 |
"epoch": 0.068,
|
| 664 |
+
"grad_norm": 6.88557243347168,
|
| 665 |
+
"kl": 0.08306531235575676,
|
| 666 |
"learning_rate": 2.4214731023046794e-07,
|
| 667 |
"loss": 0.0001,
|
| 668 |
+
"reward": 1.453125,
|
| 669 |
+
"reward_std": 0.5443578287959099,
|
| 670 |
+
"rewards/emotion_reward_func": 0.515625,
|
| 671 |
+
"rewards/format_reward_func": 0.9375,
|
| 672 |
"step": 51
|
| 673 |
},
|
| 674 |
{
|
| 675 |
+
"completion_length": 64.234375,
|
| 676 |
"epoch": 0.06933333333333333,
|
| 677 |
+
"grad_norm": 6.265974044799805,
|
| 678 |
+
"kl": 0.09051466174423695,
|
| 679 |
"learning_rate": 2.3430237011767164e-07,
|
| 680 |
"loss": 0.0001,
|
| 681 |
+
"reward": 1.484375,
|
| 682 |
+
"reward_std": 0.1875,
|
| 683 |
+
"rewards/emotion_reward_func": 0.484375,
|
| 684 |
+
"rewards/format_reward_func": 1.0,
|
| 685 |
"step": 52
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"completion_length": 55.609375,
|
| 689 |
"epoch": 0.07066666666666667,
|
| 690 |
+
"grad_norm": 5.614126205444336,
|
| 691 |
+
"kl": 0.0768609419465065,
|
| 692 |
"learning_rate": 2.264729216703714e-07,
|
| 693 |
"loss": 0.0001,
|
| 694 |
+
"reward": 1.875,
|
| 695 |
+
"reward_std": 0.31027980148792267,
|
| 696 |
+
"rewards/emotion_reward_func": 0.890625,
|
| 697 |
+
"rewards/format_reward_func": 0.984375,
|
| 698 |
"step": 53
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"completion_length": 68.859375,
|
| 702 |
"epoch": 0.072,
|
| 703 |
+
"grad_norm": 5.596243381500244,
|
| 704 |
+
"kl": 0.07597188651561737,
|
| 705 |
"learning_rate": 2.1866669160892389e-07,
|
| 706 |
"loss": 0.0001,
|
| 707 |
+
"reward": 1.125,
|
| 708 |
+
"reward_std": 0.3412870988249779,
|
| 709 |
+
"rewards/emotion_reward_func": 0.171875,
|
| 710 |
+
"rewards/format_reward_func": 0.953125,
|
| 711 |
"step": 54
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"completion_length": 51.609375,
|
| 715 |
"epoch": 0.07333333333333333,
|
| 716 |
+
"grad_norm": 7.993938446044922,
|
| 717 |
+
"kl": 0.09897390753030777,
|
| 718 |
"learning_rate": 2.1089138373994222e-07,
|
| 719 |
+
"loss": 0.0001,
|
| 720 |
+
"reward": 1.578125,
|
| 721 |
+
"reward_std": 0.47475843876600266,
|
| 722 |
+
"rewards/emotion_reward_func": 0.625,
|
| 723 |
+
"rewards/format_reward_func": 0.953125,
|
| 724 |
"step": 55
|
| 725 |
},
|
| 726 |
{
|
| 727 |
+
"completion_length": 48.265625,
|
| 728 |
"epoch": 0.07466666666666667,
|
| 729 |
+
"grad_norm": 8.550956726074219,
|
| 730 |
+
"kl": 0.10356861166656017,
|
| 731 |
"learning_rate": 2.0315467135356878e-07,
|
| 732 |
"loss": 0.0001,
|
| 733 |
+
"reward": 1.28125,
|
| 734 |
+
"reward_std": 0.3340607285499573,
|
| 735 |
+
"rewards/emotion_reward_func": 0.28125,
|
| 736 |
+
"rewards/format_reward_func": 1.0,
|
| 737 |
"step": 56
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"completion_length": 56.953125,
|
| 741 |
"epoch": 0.076,
|
| 742 |
+
"grad_norm": 6.799314498901367,
|
| 743 |
+
"kl": 0.11332254763692617,
|
| 744 |
"learning_rate": 1.954641896508644e-07,
|
| 745 |
+
"loss": 0.0001,
|
| 746 |
+
"reward": 1.671875,
|
| 747 |
+
"reward_std": 0.3965607285499573,
|
| 748 |
+
"rewards/emotion_reward_func": 0.6875,
|
| 749 |
+
"rewards/format_reward_func": 0.984375,
|
| 750 |
"step": 57
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"completion_length": 57.15625,
|
| 754 |
"epoch": 0.07733333333333334,
|
| 755 |
+
"grad_norm": 6.939026832580566,
|
| 756 |
+
"kl": 0.1864840853959322,
|
| 757 |
"learning_rate": 1.8782752820878633e-07,
|
| 758 |
+
"loss": 0.0002,
|
| 759 |
+
"reward": 1.53125,
|
| 760 |
+
"reward_std": 0.4032272547483444,
|
| 761 |
+
"rewards/emotion_reward_func": 0.609375,
|
| 762 |
+
"rewards/format_reward_func": 0.921875,
|
| 763 |
"step": 58
|
| 764 |
},
|
| 765 |
{
|
| 766 |
+
"completion_length": 50.34375,
|
| 767 |
"epoch": 0.07866666666666666,
|
| 768 |
+
"grad_norm": 7.448709487915039,
|
| 769 |
+
"kl": 0.15000650100409985,
|
| 770 |
"learning_rate": 1.802522234901927e-07,
|
| 771 |
"loss": 0.0002,
|
| 772 |
+
"reward": 1.453125,
|
| 773 |
+
"reward_std": 0.4035460501909256,
|
| 774 |
+
"rewards/emotion_reward_func": 0.46875,
|
| 775 |
+
"rewards/format_reward_func": 0.984375,
|
| 776 |
"step": 59
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"completion_length": 56.296875,
|
| 780 |
"epoch": 0.08,
|
| 781 |
+
"grad_norm": 6.004992485046387,
|
| 782 |
+
"kl": 0.08795534446835518,
|
| 783 |
"learning_rate": 1.7274575140626315e-07,
|
| 784 |
"loss": 0.0001,
|
| 785 |
+
"reward": 1.53125,
|
| 786 |
+
"reward_std": 0.4612579345703125,
|
| 787 |
+
"rewards/emotion_reward_func": 0.546875,
|
| 788 |
+
"rewards/format_reward_func": 0.984375,
|
| 789 |
"step": 60
|
| 790 |
},
|
| 791 |
{
|
| 792 |
+
"completion_length": 55.703125,
|
| 793 |
"epoch": 0.08133333333333333,
|
| 794 |
+
"grad_norm": 6.890044212341309,
|
| 795 |
+
"kl": 0.09031901508569717,
|
| 796 |
"learning_rate": 1.6531551993867715e-07,
|
| 797 |
+
"loss": 0.0001,
|
| 798 |
+
"reward": 1.453125,
|
| 799 |
+
"reward_std": 0.41576120257377625,
|
| 800 |
+
"rewards/emotion_reward_func": 0.484375,
|
| 801 |
+
"rewards/format_reward_func": 0.96875,
|
| 802 |
"step": 61
|
| 803 |
},
|
| 804 |
{
|
| 805 |
+
"completion_length": 58.625,
|
| 806 |
"epoch": 0.08266666666666667,
|
| 807 |
+
"grad_norm": 6.977153778076172,
|
| 808 |
+
"kl": 0.09828664548695087,
|
| 809 |
"learning_rate": 1.579688618288305e-07,
|
| 810 |
"loss": 0.0001,
|
| 811 |
+
"reward": 1.609375,
|
| 812 |
+
"reward_std": 0.3605812340974808,
|
| 813 |
+
"rewards/emotion_reward_func": 0.609375,
|
| 814 |
"rewards/format_reward_func": 1.0,
|
| 815 |
"step": 62
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"completion_length": 52.421875,
|
| 819 |
"epoch": 0.084,
|
| 820 |
+
"grad_norm": 3.109772205352783,
|
| 821 |
+
"kl": 0.1292141806334257,
|
| 822 |
"learning_rate": 1.5071302734130486e-07,
|
| 823 |
+
"loss": 0.0001,
|
| 824 |
+
"reward": 1.65625,
|
| 825 |
+
"reward_std": 0.15478479862213135,
|
| 826 |
+
"rewards/emotion_reward_func": 0.671875,
|
| 827 |
+
"rewards/format_reward_func": 0.984375,
|
| 828 |
"step": 63
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"completion_length": 53.578125,
|
| 832 |
"epoch": 0.08533333333333333,
|
| 833 |
+
"grad_norm": 7.28511905670166,
|
| 834 |
+
"kl": 0.1237227451056242,
|
| 835 |
"learning_rate": 1.4355517710873182e-07,
|
| 836 |
"loss": 0.0001,
|
| 837 |
+
"reward": 1.765625,
|
| 838 |
+
"reward_std": 0.5836244821548462,
|
| 839 |
+
"rewards/emotion_reward_func": 0.84375,
|
| 840 |
+
"rewards/format_reward_func": 0.921875,
|
| 841 |
"step": 64
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"completion_length": 54.265625,
|
| 845 |
"epoch": 0.08666666666666667,
|
| 846 |
+
"grad_norm": 5.760799884796143,
|
| 847 |
+
"kl": 0.10565520823001862,
|
| 848 |
"learning_rate": 1.365023750651133e-07,
|
| 849 |
"loss": 0.0001,
|
| 850 |
+
"reward": 1.59375,
|
| 851 |
+
"reward_std": 0.38039760291576385,
|
| 852 |
+
"rewards/emotion_reward_func": 0.609375,
|
| 853 |
+
"rewards/format_reward_func": 0.984375,
|
| 854 |
"step": 65
|
| 855 |
},
|
| 856 |
{
|
| 857 |
+
"completion_length": 55.9375,
|
| 858 |
"epoch": 0.088,
|
| 859 |
+
"grad_norm": 7.823127269744873,
|
| 860 |
+
"kl": 0.09790175035595894,
|
| 861 |
"learning_rate": 1.2956158147457114e-07,
|
| 862 |
"loss": 0.0001,
|
| 863 |
+
"reward": 1.625,
|
| 864 |
+
"reward_std": 0.40478479862213135,
|
| 865 |
+
"rewards/emotion_reward_func": 0.65625,
|
| 866 |
+
"rewards/format_reward_func": 0.96875,
|
| 867 |
"step": 66
|
| 868 |
},
|
| 869 |
{
|
| 870 |
+
"completion_length": 44.671875,
|
| 871 |
"epoch": 0.08933333333333333,
|
| 872 |
+
"grad_norm": 4.364702224731445,
|
| 873 |
+
"kl": 0.11339546367526054,
|
| 874 |
"learning_rate": 1.2273964606240718e-07,
|
| 875 |
"loss": 0.0001,
|
| 876 |
+
"reward": 1.34375,
|
| 877 |
+
"reward_std": 0.18217839300632477,
|
| 878 |
+
"rewards/emotion_reward_func": 0.34375,
|
| 879 |
+
"rewards/format_reward_func": 1.0,
|
| 880 |
"step": 67
|
| 881 |
},
|
| 882 |
{
|
| 883 |
+
"completion_length": 57.421875,
|
| 884 |
"epoch": 0.09066666666666667,
|
| 885 |
+
"grad_norm": 5.267661094665527,
|
| 886 |
+
"kl": 0.10790612921118736,
|
| 887 |
"learning_rate": 1.1604330125525078e-07,
|
| 888 |
"loss": 0.0001,
|
| 889 |
+
"reward": 1.1875,
|
| 890 |
+
"reward_std": 0.2545653209090233,
|
| 891 |
+
"rewards/emotion_reward_func": 0.21875,
|
| 892 |
+
"rewards/format_reward_func": 0.96875,
|
| 893 |
"step": 68
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"completion_length": 48.921875,
|
| 897 |
"epoch": 0.092,
|
| 898 |
+
"grad_norm": 5.409526824951172,
|
| 899 |
+
"kl": 0.11567417718470097,
|
| 900 |
"learning_rate": 1.0947915553696741e-07,
|
| 901 |
"loss": 0.0001,
|
| 902 |
+
"reward": 1.453125,
|
| 903 |
+
"reward_std": 0.23328250646591187,
|
| 904 |
+
"rewards/emotion_reward_func": 0.453125,
|
| 905 |
+
"rewards/format_reward_func": 1.0,
|
| 906 |
"step": 69
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"completion_length": 55.828125,
|
| 910 |
"epoch": 0.09333333333333334,
|
| 911 |
+
"grad_norm": 7.669867992401123,
|
| 912 |
+
"kl": 3.9372363202273846,
|
| 913 |
"learning_rate": 1.0305368692688174e-07,
|
| 914 |
+
"loss": 0.0039,
|
| 915 |
+
"reward": 1.34375,
|
| 916 |
+
"reward_std": 0.3306438848376274,
|
| 917 |
+
"rewards/emotion_reward_func": 0.359375,
|
| 918 |
+
"rewards/format_reward_func": 0.984375,
|
| 919 |
"step": 70
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"completion_length": 43.125,
|
| 923 |
"epoch": 0.09466666666666666,
|
| 924 |
+
"grad_norm": 8.619794845581055,
|
| 925 |
+
"kl": 0.13703159242868423,
|
| 926 |
"learning_rate": 9.677323658675593e-08,
|
| 927 |
"loss": 0.0001,
|
| 928 |
+
"reward": 1.5625,
|
| 929 |
+
"reward_std": 0.3234764039516449,
|
| 930 |
+
"rewards/emotion_reward_func": 0.578125,
|
| 931 |
+
"rewards/format_reward_func": 0.984375,
|
| 932 |
"step": 71
|
| 933 |
},
|
| 934 |
{
|
| 935 |
+
"completion_length": 47.484375,
|
| 936 |
"epoch": 0.096,
|
| 937 |
+
"grad_norm": 5.2266740798950195,
|
| 938 |
+
"kl": 0.1516919508576393,
|
| 939 |
"learning_rate": 9.064400256282755e-08,
|
| 940 |
+
"loss": 0.0002,
|
| 941 |
+
"reward": 1.296875,
|
| 942 |
+
"reward_std": 0.34940721094608307,
|
| 943 |
+
"rewards/emotion_reward_func": 0.3125,
|
| 944 |
+
"rewards/format_reward_func": 0.984375,
|
| 945 |
"step": 72
|
| 946 |
},
|
| 947 |
{
|
| 948 |
+
"completion_length": 60.5,
|
| 949 |
"epoch": 0.09733333333333333,
|
| 950 |
+
"grad_norm": 6.890865325927734,
|
| 951 |
+
"kl": 0.10698151029646397,
|
| 952 |
"learning_rate": 8.467203366908707e-08,
|
| 953 |
"loss": 0.0001,
|
| 954 |
+
"reward": 1.375,
|
| 955 |
+
"reward_std": 0.3380199372768402,
|
| 956 |
+
"rewards/emotion_reward_func": 0.40625,
|
| 957 |
+
"rewards/format_reward_func": 0.96875,
|
| 958 |
"step": 73
|
| 959 |
},
|
| 960 |
{
|
| 961 |
+
"completion_length": 45.8125,
|
| 962 |
"epoch": 0.09866666666666667,
|
| 963 |
+
"grad_norm": 4.76033878326416,
|
| 964 |
+
"kl": 0.2001918088644743,
|
| 965 |
"learning_rate": 7.886322351782782e-08,
|
| 966 |
+
"loss": 0.0002,
|
| 967 |
+
"reward": 1.6875,
|
| 968 |
+
"reward_std": 0.26933756470680237,
|
| 969 |
+
"rewards/emotion_reward_func": 0.703125,
|
| 970 |
+
"rewards/format_reward_func": 0.984375,
|
| 971 |
"step": 74
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"completion_length": 53.640625,
|
| 975 |
"epoch": 0.1,
|
| 976 |
+
"grad_norm": 7.104291915893555,
|
| 977 |
+
"kl": 0.09334802068769932,
|
| 978 |
"learning_rate": 7.322330470336313e-08,
|
| 979 |
"loss": 0.0001,
|
| 980 |
+
"reward": 1.453125,
|
| 981 |
+
"reward_std": 0.3290881961584091,
|
| 982 |
+
"rewards/emotion_reward_func": 0.484375,
|
| 983 |
+
"rewards/format_reward_func": 0.96875,
|
| 984 |
"step": 75
|
| 985 |
},
|
| 986 |
{
|
| 987 |
+
"completion_length": 45.109375,
|
| 988 |
"epoch": 0.10133333333333333,
|
| 989 |
+
"grad_norm": 10.217005729675293,
|
| 990 |
+
"kl": 0.12721979059278965,
|
| 991 |
"learning_rate": 6.775784314464716e-08,
|
| 992 |
"loss": 0.0001,
|
| 993 |
+
"reward": 1.4375,
|
| 994 |
+
"reward_std": 0.3354101926088333,
|
| 995 |
+
"rewards/emotion_reward_func": 0.4375,
|
| 996 |
+
"rewards/format_reward_func": 1.0,
|
| 997 |
"step": 76
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
+
"completion_length": 49.390625,
|
| 1001 |
"epoch": 0.10266666666666667,
|
| 1002 |
+
"grad_norm": 5.090665817260742,
|
| 1003 |
+
"kl": 0.1952703520655632,
|
| 1004 |
"learning_rate": 6.24722325923851e-08,
|
| 1005 |
"loss": 0.0002,
|
| 1006 |
+
"reward": 1.5625,
|
| 1007 |
+
"reward_std": 0.21039125323295593,
|
| 1008 |
+
"rewards/emotion_reward_func": 0.5625,
|
| 1009 |
"rewards/format_reward_func": 1.0,
|
| 1010 |
"step": 77
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
+
"completion_length": 40.953125,
|
| 1014 |
"epoch": 0.104,
|
| 1015 |
+
"grad_norm": 10.18713665008545,
|
| 1016 |
+
"kl": 0.1544223688542843,
|
| 1017 |
"learning_rate": 5.737168930605271e-08,
|
| 1018 |
"loss": 0.0002,
|
| 1019 |
+
"reward": 1.609375,
|
| 1020 |
+
"reward_std": 0.44701361656188965,
|
| 1021 |
+
"rewards/emotion_reward_func": 0.640625,
|
| 1022 |
"rewards/format_reward_func": 0.96875,
|
| 1023 |
"step": 78
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"completion_length": 45.84375,
|
| 1027 |
"epoch": 0.10533333333333333,
|
| 1028 |
+
"grad_norm": 6.143667221069336,
|
| 1029 |
+
"kl": 0.14481117203831673,
|
| 1030 |
"learning_rate": 5.246124690607739e-08,
|
| 1031 |
+
"loss": 0.0001,
|
| 1032 |
+
"reward": 1.515625,
|
| 1033 |
+
"reward_std": 0.2715607285499573,
|
| 1034 |
+
"rewards/emotion_reward_func": 0.515625,
|
| 1035 |
+
"rewards/format_reward_func": 1.0,
|
| 1036 |
"step": 79
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
+
"completion_length": 46.59375,
|
| 1040 |
"epoch": 0.10666666666666667,
|
| 1041 |
+
"grad_norm": 5.600602149963379,
|
| 1042 |
+
"kl": 0.10117345489561558,
|
| 1043 |
"learning_rate": 4.774575140626316e-08,
|
| 1044 |
+
"loss": 0.0001,
|
| 1045 |
+
"reward": 1.890625,
|
| 1046 |
+
"reward_std": 0.1280868798494339,
|
| 1047 |
+
"rewards/emotion_reward_func": 0.890625,
|
| 1048 |
+
"rewards/format_reward_func": 1.0,
|
| 1049 |
"step": 80
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
+
"completion_length": 55.265625,
|
| 1053 |
"epoch": 0.108,
|
| 1054 |
+
"grad_norm": 8.253979682922363,
|
| 1055 |
+
"kl": 0.11770397983491421,
|
| 1056 |
"learning_rate": 4.3229856431359513e-08,
|
| 1057 |
+
"loss": 0.0001,
|
| 1058 |
+
"reward": 1.515625,
|
| 1059 |
+
"reward_std": 0.5620162785053253,
|
| 1060 |
+
"rewards/emotion_reward_func": 0.5625,
|
| 1061 |
+
"rewards/format_reward_func": 0.953125,
|
| 1062 |
"step": 81
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
+
"completion_length": 55.34375,
|
| 1066 |
"epoch": 0.10933333333333334,
|
| 1067 |
+
"grad_norm": 7.1831135749816895,
|
| 1068 |
+
"kl": 0.1115355659276247,
|
| 1069 |
"learning_rate": 3.8918018624496286e-08,
|
| 1070 |
"loss": 0.0001,
|
| 1071 |
+
"reward": 1.375,
|
| 1072 |
+
"reward_std": 0.3838024437427521,
|
| 1073 |
+
"rewards/emotion_reward_func": 0.390625,
|
| 1074 |
+
"rewards/format_reward_func": 0.984375,
|
| 1075 |
"step": 82
|
| 1076 |
},
|
| 1077 |
{
|
| 1078 |
+
"completion_length": 48.40625,
|
| 1079 |
"epoch": 0.11066666666666666,
|
| 1080 |
+
"grad_norm": 7.065600395202637,
|
| 1081 |
+
"kl": 0.14161021262407303,
|
| 1082 |
"learning_rate": 3.481449324901411e-08,
|
| 1083 |
"loss": 0.0001,
|
| 1084 |
+
"reward": 1.515625,
|
| 1085 |
+
"reward_std": 0.3940916210412979,
|
| 1086 |
+
"rewards/emotion_reward_func": 0.546875,
|
| 1087 |
"rewards/format_reward_func": 0.96875,
|
| 1088 |
"step": 83
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
+
"completion_length": 52.5625,
|
| 1092 |
"epoch": 0.112,
|
| 1093 |
+
"grad_norm": 7.827910423278809,
|
| 1094 |
+
"kl": 0.16811930015683174,
|
| 1095 |
"learning_rate": 3.092332998903416e-08,
|
| 1096 |
+
"loss": 0.0002,
|
| 1097 |
+
"reward": 1.5,
|
| 1098 |
+
"reward_std": 0.5367372632026672,
|
| 1099 |
+
"rewards/emotion_reward_func": 0.546875,
|
| 1100 |
+
"rewards/format_reward_func": 0.953125,
|
| 1101 |
"step": 84
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
+
"completion_length": 51.328125,
|
| 1105 |
"epoch": 0.11333333333333333,
|
| 1106 |
+
"grad_norm": 4.83447265625,
|
| 1107 |
+
"kl": 0.11310878023505211,
|
| 1108 |
"learning_rate": 2.724836895290805e-08,
|
| 1109 |
"loss": 0.0001,
|
| 1110 |
+
"reward": 1.640625,
|
| 1111 |
+
"reward_std": 0.1280868798494339,
|
| 1112 |
+
"rewards/emotion_reward_func": 0.640625,
|
| 1113 |
"rewards/format_reward_func": 1.0,
|
| 1114 |
"step": 85
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
+
"completion_length": 47.578125,
|
| 1118 |
"epoch": 0.11466666666666667,
|
| 1119 |
+
"grad_norm": 5.901933670043945,
|
| 1120 |
+
"kl": 0.1336957048624754,
|
| 1121 |
"learning_rate": 2.379323688349516e-08,
|
| 1122 |
"loss": 0.0001,
|
| 1123 |
+
"reward": 1.375,
|
| 1124 |
+
"reward_std": 0.21039125323295593,
|
| 1125 |
+
"rewards/emotion_reward_func": 0.375,
|
| 1126 |
"rewards/format_reward_func": 1.0,
|
| 1127 |
"step": 86
|
| 1128 |
},
|
| 1129 |
{
|
| 1130 |
+
"completion_length": 50.53125,
|
| 1131 |
"epoch": 0.116,
|
| 1132 |
+
"grad_norm": 5.392825603485107,
|
| 1133 |
+
"kl": 0.1246971283107996,
|
| 1134 |
"learning_rate": 2.0561343579004715e-08,
|
| 1135 |
"loss": 0.0001,
|
| 1136 |
+
"reward": 1.6875,
|
| 1137 |
+
"reward_std": 0.3024514466524124,
|
| 1138 |
+
"rewards/emotion_reward_func": 0.71875,
|
| 1139 |
+
"rewards/format_reward_func": 0.96875,
|
| 1140 |
"step": 87
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
+
"completion_length": 52.515625,
|
| 1144 |
"epoch": 0.11733333333333333,
|
| 1145 |
+
"grad_norm": 7.34331750869751,
|
| 1146 |
+
"kl": 0.11970937624573708,
|
| 1147 |
"learning_rate": 1.7555878527937163e-08,
|
| 1148 |
+
"loss": 0.0001,
|
| 1149 |
+
"reward": 1.34375,
|
| 1150 |
+
"reward_std": 0.40886765718460083,
|
| 1151 |
+
"rewards/emotion_reward_func": 0.359375,
|
| 1152 |
+
"rewards/format_reward_func": 0.984375,
|
| 1153 |
"step": 88
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
+
"completion_length": 58.03125,
|
| 1157 |
"epoch": 0.11866666666666667,
|
| 1158 |
+
"grad_norm": 4.543256759643555,
|
| 1159 |
+
"kl": 0.12226046249270439,
|
| 1160 |
"learning_rate": 1.4779807761443635e-08,
|
| 1161 |
"loss": 0.0001,
|
| 1162 |
+
"reward": 1.546875,
|
| 1163 |
+
"reward_std": 0.3359370082616806,
|
| 1164 |
+
"rewards/emotion_reward_func": 0.578125,
|
| 1165 |
+
"rewards/format_reward_func": 0.96875,
|
| 1166 |
"step": 89
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
+
"completion_length": 52.265625,
|
| 1170 |
"epoch": 0.12,
|
| 1171 |
+
"grad_norm": 7.780374050140381,
|
| 1172 |
+
"kl": 0.1240895576775074,
|
| 1173 |
"learning_rate": 1.2235870926211616e-08,
|
| 1174 |
"loss": 0.0001,
|
| 1175 |
+
"reward": 1.59375,
|
| 1176 |
+
"reward_std": 0.25,
|
| 1177 |
+
"rewards/emotion_reward_func": 0.59375,
|
| 1178 |
+
"rewards/format_reward_func": 1.0,
|
| 1179 |
"step": 90
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
+
"completion_length": 52.5,
|
| 1183 |
"epoch": 0.12133333333333333,
|
| 1184 |
+
"grad_norm": 3.1433615684509277,
|
| 1185 |
+
"kl": 0.10472103767096996,
|
| 1186 |
"learning_rate": 9.926578580764234e-09,
|
| 1187 |
"loss": 0.0001,
|
| 1188 |
+
"reward": 1.734375,
|
| 1189 |
+
"reward_std": 0.0625,
|
| 1190 |
+
"rewards/emotion_reward_func": 0.734375,
|
| 1191 |
+
"rewards/format_reward_func": 1.0,
|
| 1192 |
"step": 91
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"completion_length": 50.921875,
|
| 1196 |
"epoch": 0.12266666666666666,
|
| 1197 |
+
"grad_norm": 7.484493732452393,
|
| 1198 |
+
"kl": 0.10997971147298813,
|
| 1199 |
"learning_rate": 7.85420971784223e-09,
|
| 1200 |
"loss": 0.0001,
|
| 1201 |
+
"reward": 1.640625,
|
| 1202 |
+
"reward_std": 0.2829566150903702,
|
| 1203 |
+
"rewards/emotion_reward_func": 0.640625,
|
| 1204 |
+
"rewards/format_reward_func": 1.0,
|
| 1205 |
"step": 92
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
+
"completion_length": 52.015625,
|
| 1209 |
"epoch": 0.124,
|
| 1210 |
+
"grad_norm": 8.367355346679688,
|
| 1211 |
+
"kl": 0.1366768404841423,
|
| 1212 |
"learning_rate": 6.020809515313141e-09,
|
| 1213 |
"loss": 0.0001,
|
| 1214 |
+
"reward": 1.28125,
|
| 1215 |
+
"reward_std": 0.3538651019334793,
|
| 1216 |
+
"rewards/emotion_reward_func": 0.28125,
|
| 1217 |
+
"rewards/format_reward_func": 1.0,
|
| 1218 |
"step": 93
|
| 1219 |
},
|
| 1220 |
{
|
| 1221 |
+
"completion_length": 45.796875,
|
| 1222 |
"epoch": 0.12533333333333332,
|
| 1223 |
+
"grad_norm": 6.152498245239258,
|
| 1224 |
+
"kl": 0.1532982587814331,
|
| 1225 |
"learning_rate": 4.4281873178278475e-09,
|
| 1226 |
"loss": 0.0002,
|
| 1227 |
+
"reward": 1.640625,
|
| 1228 |
+
"reward_std": 0.414583221077919,
|
| 1229 |
+
"rewards/emotion_reward_func": 0.671875,
|
| 1230 |
+
"rewards/format_reward_func": 0.96875,
|
| 1231 |
"step": 94
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
+
"completion_length": 50.25,
|
| 1235 |
"epoch": 0.12666666666666668,
|
| 1236 |
+
"grad_norm": 0.020403465256094933,
|
| 1237 |
+
"kl": 0.11631555296480656,
|
| 1238 |
"learning_rate": 3.077914851215585e-09,
|
| 1239 |
+
"loss": 0.0001,
|
| 1240 |
+
"reward": 1.5,
|
| 1241 |
+
"reward_std": 0.0,
|
| 1242 |
+
"rewards/emotion_reward_func": 0.5,
|
| 1243 |
+
"rewards/format_reward_func": 1.0,
|
| 1244 |
"step": 95
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
+
"completion_length": 53.671875,
|
| 1248 |
"epoch": 0.128,
|
| 1249 |
+
"grad_norm": 6.355586528778076,
|
| 1250 |
+
"kl": 0.11114749684929848,
|
| 1251 |
"learning_rate": 1.9713246713805587e-09,
|
| 1252 |
"loss": 0.0001,
|
| 1253 |
+
"reward": 1.578125,
|
| 1254 |
+
"reward_std": 0.33949775248765945,
|
| 1255 |
+
"rewards/emotion_reward_func": 0.59375,
|
| 1256 |
+
"rewards/format_reward_func": 0.984375,
|
| 1257 |
"step": 96
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
+
"completion_length": 51.921875,
|
| 1261 |
"epoch": 0.12933333333333333,
|
| 1262 |
+
"grad_norm": 7.896821022033691,
|
| 1263 |
+
"kl": 0.14222519844770432,
|
| 1264 |
"learning_rate": 1.1095088492300008e-09,
|
| 1265 |
+
"loss": 0.0001,
|
| 1266 |
+
"reward": 1.625,
|
| 1267 |
+
"reward_std": 0.1905868798494339,
|
| 1268 |
+
"rewards/emotion_reward_func": 0.625,
|
| 1269 |
+
"rewards/format_reward_func": 1.0,
|
| 1270 |
"step": 97
|
| 1271 |
},
|
| 1272 |
{
|
| 1273 |
+
"completion_length": 52.90625,
|
| 1274 |
"epoch": 0.13066666666666665,
|
| 1275 |
+
"grad_norm": 8.117393493652344,
|
| 1276 |
+
"kl": 0.29374901205301285,
|
| 1277 |
"learning_rate": 4.933178929321102e-10,
|
| 1278 |
+
"loss": 0.0003,
|
| 1279 |
+
"reward": 1.65625,
|
| 1280 |
+
"reward_std": 0.5836337506771088,
|
| 1281 |
+
"rewards/emotion_reward_func": 0.71875,
|
| 1282 |
+
"rewards/format_reward_func": 0.9375,
|
| 1283 |
"step": 98
|
| 1284 |
},
|
| 1285 |
{
|
| 1286 |
+
"completion_length": 49.234375,
|
| 1287 |
"epoch": 0.132,
|
| 1288 |
+
"grad_norm": 4.081859111785889,
|
| 1289 |
+
"kl": 0.12134127877652645,
|
| 1290 |
"learning_rate": 1.2335990856709998e-10,
|
| 1291 |
"loss": 0.0001,
|
| 1292 |
+
"reward": 1.75,
|
| 1293 |
+
"reward_std": 0.2367546260356903,
|
| 1294 |
+
"rewards/emotion_reward_func": 0.765625,
|
| 1295 |
+
"rewards/format_reward_func": 0.984375,
|
| 1296 |
"step": 99
|
| 1297 |
},
|
| 1298 |
{
|
| 1299 |
+
"completion_length": 51.25,
|
| 1300 |
"epoch": 0.13333333333333333,
|
| 1301 |
+
"grad_norm": 6.719541072845459,
|
| 1302 |
+
"kl": 0.1454935073852539,
|
| 1303 |
"learning_rate": 0.0,
|
| 1304 |
"loss": 0.0001,
|
| 1305 |
+
"reward": 1.609375,
|
| 1306 |
+
"reward_std": 0.36967839300632477,
|
| 1307 |
+
"rewards/emotion_reward_func": 0.640625,
|
| 1308 |
+
"rewards/format_reward_func": 0.96875,
|
| 1309 |
"step": 100
|
| 1310 |
}
|
| 1311 |
],
|
|
|
|
| 1327 |
}
|
| 1328 |
},
|
| 1329 |
"total_flos": 0.0,
|
| 1330 |
+
"train_batch_size": 4,
|
| 1331 |
"trial_name": null,
|
| 1332 |
"trial_params": null
|
| 1333 |
}
|
checkpoint-100/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc6e728504ceeed070362223f2228cf2aa1bf386361fcc0c0d4d877e62196c58
|
| 3 |
+
size 5752
|