IDoNotHaveAName/X-R1-3epoch
06
1{2 "best_metric": null,3 "best_model_checkpoint": null,4 "epoch": 2.992,5 "eval_steps": 10,6 "global_step": 279,7 "is_hyper_param_search": false,8 "is_local_process_zero": true,9 "is_world_process_zero": true,10 "log_history": [11 {12 "clip_ratio": 0.0,13 "completion_length": 410.9166774749756,14 "epoch": 0.010666666666666666,15 "grad_norm": 0.5299306511878967,16 "kl": 0.0,17 "learning_rate": 1.0714285714285714e-07,18 "loss": 0.0126,19 "reward": 0.7291666772216558,20 "reward_std": 0.1530931107699871,21 "rewards/accuracy_reward": 0.7291666772216558,22 "rewards/format_reward": 0.0,23 "step": 124 },25 {26 "clip_ratio": 0.0,27 "completion_length": 422.85417556762695,28 "epoch": 0.021333333333333333,29 "grad_norm": 0.4661875367164612,30 "kl": 0.0,31 "learning_rate": 2.1428571428571428e-07,32 "loss": 0.0183,33 "reward": 0.7916666753590107,34 "reward_std": 0.24859581887722015,35 "rewards/accuracy_reward": 0.7916666753590107,36 "rewards/format_reward": 0.0,37 "step": 238 },39 {40 "clip_ratio": 0.0,41 "completion_length": 438.2708549499512,42 "epoch": 0.032,43 "grad_norm": 0.8872589468955994,44 "kl": -8.471310138702393e-06,45 "learning_rate": 3.214285714285714e-07,46 "loss": 0.0187,47 "reward": 0.8958333432674408,48 "reward_std": 0.1705273948609829,49 "rewards/accuracy_reward": 0.8958333432674408,50 "rewards/format_reward": 0.0,51 "step": 352 },53 {54 "clip_ratio": 0.0,55 "completion_length": 436.6666736602783,56 "epoch": 0.042666666666666665,57 "grad_norm": 0.09845240414142609,58 "kl": -7.5623393058776855e-06,59 "learning_rate": 4.2857142857142857e-07,60 "loss": 0.0108,61 "reward": 0.7916666734963655,62 "reward_std": 0.16661179810762405,63 "rewards/accuracy_reward": 0.7916666734963655,64 "rewards/format_reward": 0.0,65 "step": 466 },67 {68 "clip_ratio": 0.0,69 "completion_length": 416.70834732055664,70 "epoch": 0.05333333333333334,71 "grad_norm": 0.565118134021759,72 "kl": -5.11854887008667e-06,73 "learning_rate": 5.357142857142858e-07,74 "loss": 0.0196,75 "reward": 0.9583333358168602,76 "reward_std": 0.06454972922801971,77 "rewards/accuracy_reward": 0.9583333358168602,78 "rewards/format_reward": 0.0,79 "step": 580 },81 {82 "clip_ratio": 0.0,83 "completion_length": 430.27084732055664,84 "epoch": 0.064,85 "grad_norm": 0.9171705842018127,86 "kl": 1.4778226613998413e-05,87 "learning_rate": 6.428571428571428e-07,88 "loss": 0.0091,89 "reward": 0.8750000111758709,90 "reward_std": 0.16661179810762405,91 "rewards/accuracy_reward": 0.8750000111758709,92 "rewards/format_reward": 0.0,93 "step": 694 },95 {96 "clip_ratio": 0.0,97 "completion_length": 400.00000762939453,98 "epoch": 0.07466666666666667,99 "grad_norm": 0.5704061388969421,100 "kl": -8.910894393920898e-06,101 "learning_rate": 7.5e-07,102 "loss": -0.0,103 "reward": 0.9166666679084301,104 "reward_std": 0.06454972922801971,105 "rewards/accuracy_reward": 0.9166666679084301,106 "rewards/format_reward": 0.0,107 "step": 7108 },109 {110 "clip_ratio": 0.0,111 "completion_length": 342.5833396911621,112 "epoch": 0.08533333333333333,113 "grad_norm": 0.4634876251220703,114 "kl": 3.3229589462280273e-06,115 "learning_rate": 8.571428571428571e-07,116 "loss": 0.0097,117 "reward": 0.9166666716337204,118 "reward_std": 0.12909945845603943,119 "rewards/accuracy_reward": 0.9166666716337204,120 "rewards/format_reward": 0.0,121 "step": 8122 },123 {124 "clip_ratio": 0.0,125 "completion_length": 446.02084159851074,126 "epoch": 0.096,127 "grad_norm": 0.4667946994304657,128 "kl": 5.4001808166503906e-05,129 "learning_rate": 9.642857142857145e-07,130 "loss": 0.005,131 "reward": 0.7708333358168602,132 "reward_std": 0.13301505148410797,133 "rewards/accuracy_reward": 0.7708333358168602,134 "rewards/format_reward": 0.0,135 "step": 9136 },137 {138 "clip_ratio": 0.0,139 "completion_length": 410.5416831970215,140 "epoch": 0.10666666666666667,141 "grad_norm": 0.0004501362855080515,142 "kl": 1.0106712579727173e-05,143 "learning_rate": 1.0714285714285716e-06,144 "loss": 0.0,145 "reward": 0.875,146 "reward_std": 0.0,147 "rewards/accuracy_reward": 0.875,148 "rewards/format_reward": 0.0,149 "step": 10150 },151 {152 "clip_ratio": 0.0,153 "completion_length": 337.3333377838135,154 "epoch": 0.11733333333333333,155 "grad_norm": 0.0002833263133652508,156 "kl": 1.4334917068481445e-05,157 "learning_rate": 1.1785714285714285e-06,158 "loss": 0.0,159 "reward": 1.0,160 "reward_std": 0.0,161 "rewards/accuracy_reward": 1.0,162 "rewards/format_reward": 0.0,163 "step": 11164 },165 {166 "clip_ratio": 0.0,167 "completion_length": 459.43750762939453,168 "epoch": 0.128,169 "grad_norm": 1.180604338645935,170 "kl": 0.0019370466470718384,171 "learning_rate": 1.2857142857142856e-06,172 "loss": 0.0508,173 "reward": 0.7291666697710752,174 "reward_std": 0.1801304966211319,175 "rewards/accuracy_reward": 0.7291666697710752,176 "rewards/format_reward": 0.0,177 "step": 12178 },179 {180 "clip_ratio": 0.0,181 "completion_length": 333.41667556762695,182 "epoch": 0.13866666666666666,183 "grad_norm": 0.06026441603899002,184 "kl": -1.620757393538952e-05,185 "learning_rate": 1.392857142857143e-06,186 "loss": -0.0034,187 "reward": 0.895833333954215,188 "reward_std": 0.05103103816509247,189 "rewards/accuracy_reward": 0.895833333954215,190 "rewards/format_reward": 0.0,191 "step": 13192 },193 {194 "clip_ratio": 0.0,195 "completion_length": 326.9166774749756,196 "epoch": 0.14933333333333335,197 "grad_norm": 0.03230924531817436,198 "kl": -2.1424144506454468e-05,199 "learning_rate": 1.5e-06,200 "loss": -0.0172,201 "reward": 0.9583333432674408,202 "reward_std": 0.10206207260489464,203 "rewards/accuracy_reward": 0.9583333432674408,204 "rewards/format_reward": 0.0,205 "step": 14206 },207 {208 "clip_ratio": 0.0,209 "completion_length": 446.2500114440918,210 "epoch": 0.16,211 "grad_norm": 0.8342331051826477,212 "kl": 0.005100443959236145,213 "learning_rate": 1.6071428571428572e-06,214 "loss": 0.0341,215 "reward": 0.7708333395421505,216 "reward_std": 0.11558076366782188,217 "rewards/accuracy_reward": 0.7708333395421505,218 "rewards/format_reward": 0.0,219 "step": 15220 },221 {222 "clip_ratio": 0.0,223 "completion_length": 426.06251525878906,224 "epoch": 0.17066666666666666,225 "grad_norm": 0.014114960096776485,226 "kl": 0.0043880753219127655,227 "learning_rate": 1.7142857142857143e-06,228 "loss": 0.0003,229 "reward": 0.625,230 "reward_std": 0.0,231 "rewards/accuracy_reward": 0.625,232 "rewards/format_reward": 0.0,233 "step": 16234 },235 {236 "clip_ratio": 0.0,237 "completion_length": 471.20834732055664,238 "epoch": 0.18133333333333335,239 "grad_norm": 0.015604740008711815,240 "kl": 0.007741432636976242,241 "learning_rate": 1.8214285714285714e-06,242 "loss": 0.0003,243 "reward": 0.75,244 "reward_std": 0.0,245 "rewards/accuracy_reward": 0.75,246 "rewards/format_reward": 0.0,247 "step": 17248 },249 {250 "clip_ratio": 0.0,251 "completion_length": 458.2708511352539,252 "epoch": 0.192,253 "grad_norm": 0.45530879497528076,254 "kl": 0.021405041217803955,255 "learning_rate": 1.928571428571429e-06,256 "loss": 0.018,257 "reward": 0.7708333358168602,258 "reward_std": 0.13301505148410797,259 "rewards/accuracy_reward": 0.7708333358168602,260 "rewards/format_reward": 0.0,261 "step": 18262 },263 {264 "clip_ratio": 0.0,265 "completion_length": 307.43750762939453,266 "epoch": 0.20266666666666666,267 "grad_norm": 0.0004259841807652265,268 "kl": 2.8014183044433594e-05,269 "learning_rate": 2.035714285714286e-06,270 "loss": 0.0,271 "reward": 1.0,272 "reward_std": 0.0,273 "rewards/accuracy_reward": 1.0,274 "rewards/format_reward": 0.0,275 "step": 19276 },277 {278 "clip_ratio": 0.0,279 "completion_length": 529.1458492279053,280 "epoch": 0.21333333333333335,281 "grad_norm": 1.0350282192230225,282 "kl": 0.023723304271697998,283 "learning_rate": 2.142857142857143e-06,284 "loss": 0.0356,285 "reward": 0.5625000055879354,286 "reward_std": 0.1705273911356926,287 "rewards/accuracy_reward": 0.5625000055879354,288 "rewards/format_reward": 0.0,289 "step": 20290 },291 {292 "clip_ratio": 0.0,293 "completion_length": 365.87500953674316,294 "epoch": 0.224,295 "grad_norm": 0.01775968447327614,296 "kl": 0.00790809583850205,297 "learning_rate": 2.25e-06,298 "loss": 0.0005,299 "reward": 0.875,300 "reward_std": 0.0,301 "rewards/accuracy_reward": 0.875,302 "rewards/format_reward": 0.0,303 "step": 21304 },305 {306 "clip_ratio": 0.0,307 "completion_length": 454.8958377838135,308 "epoch": 0.23466666666666666,309 "grad_norm": 0.05057643726468086,310 "kl": 0.026109017431735992,311 "learning_rate": 2.357142857142857e-06,312 "loss": 0.0149,313 "reward": 0.7708333376795053,314 "reward_std": 0.1801304928958416,315 "rewards/accuracy_reward": 0.7708333376795053,316 "rewards/format_reward": 0.0,317 "step": 22318 },319 {320 "clip_ratio": 0.0,321 "completion_length": 450.50000762939453,322 "epoch": 0.24533333333333332,323 "grad_norm": 0.1461334526538849,324 "kl": 0.0741237998008728,325 "learning_rate": 2.4642857142857143e-06,326 "loss": -0.0041,327 "reward": 0.7500000055879354,328 "reward_std": 0.23899271711707115,329 "rewards/accuracy_reward": 0.7500000055879354,330 "rewards/format_reward": 0.0,331 "step": 23332 },333 {334 "clip_ratio": 0.0,335 "completion_length": 387.18750762939453,336 "epoch": 0.256,337 "grad_norm": 0.5979704260826111,338 "kl": 0.022205054759979248,339 "learning_rate": 2.571428571428571e-06,340 "loss": 0.0201,341 "reward": 0.8541666716337204,342 "reward_std": 0.05103103443980217,343 "rewards/accuracy_reward": 0.8541666716337204,344 "rewards/format_reward": 0.0,345 "step": 24346 },347 {348 "clip_ratio": 0.0,349 "completion_length": 500.43750762939453,350 "epoch": 0.26666666666666666,351 "grad_norm": 0.7714133262634277,352 "kl": 0.6279669404029846,353 "learning_rate": 2.678571428571429e-06,354 "loss": 0.1006,355 "reward": 0.5000000055879354,356 "reward_std": 0.10206207260489464,357 "rewards/accuracy_reward": 0.5000000055879354,358 "rewards/format_reward": 0.0,359 "step": 25360 },361 {362 "clip_ratio": 0.0,363 "completion_length": 489.0833435058594,364 "epoch": 0.2773333333333333,365 "grad_norm": 0.1741541475057602,366 "kl": 0.5411486104130745,367 "learning_rate": 2.785714285714286e-06,368 "loss": 0.0444,369 "reward": 0.8541666716337204,370 "reward_std": 0.05103103443980217,371 "rewards/accuracy_reward": 0.8541666716337204,372 "rewards/format_reward": 0.0,373 "step": 26374 },375 {376 "clip_ratio": 0.0,377 "completion_length": 483.9375190734863,378 "epoch": 0.288,379 "grad_norm": 0.11233610659837723,380 "kl": 0.18515753000974655,381 "learning_rate": 2.892857142857143e-06,382 "loss": 0.0476,383 "reward": 0.8125,384 "reward_std": 0.06846532225608826,385 "rewards/accuracy_reward": 0.8125,386 "rewards/format_reward": 0.0,387 "step": 27388 },389 {390 "clip_ratio": 0.0,391 "completion_length": 411.14584732055664,392 "epoch": 0.2986666666666667,393 "grad_norm": 0.0639011412858963,394 "kl": 0.11302819848060608,395 "learning_rate": 3e-06,396 "loss": 0.0116,397 "reward": 0.8333333358168602,398 "reward_std": 0.06454972922801971,399 "rewards/accuracy_reward": 0.8333333358168602,400 "rewards/format_reward": 0.0,401 "step": 28402 },403 {404 "clip_ratio": 0.0,405 "completion_length": 398.83334732055664,406 "epoch": 0.30933333333333335,407 "grad_norm": 0.020812369883060455,408 "kl": 0.03338313102722168,409 "learning_rate": 2.999882508108334e-06,410 "loss": 0.0026,411 "reward": 0.9166666679084301,412 "reward_std": 0.06454972922801971,413 "rewards/accuracy_reward": 0.9166666679084301,414 "rewards/format_reward": 0.0,415 "step": 29416 },417 {418 "clip_ratio": 0.0,419 "completion_length": 409.3125114440918,420 "epoch": 0.32,421 "grad_norm": 0.5567188262939453,422 "kl": 0.6645529866218567,423 "learning_rate": 2.9995300508391286e-06,424 "loss": 0.0308,425 "reward": 0.875,426 "reward_std": 0.0,427 "rewards/accuracy_reward": 0.875,428 "rewards/format_reward": 0.0,429 "step": 30430 },431 {432 "clip_ratio": 0.0,433 "completion_length": 435.50001525878906,434 "epoch": 0.33066666666666666,435 "grad_norm": 0.3817005455493927,436 "kl": 0.3913140296936035,437 "learning_rate": 2.998942683406879e-06,438 "loss": 0.0262,439 "reward": 0.7291666716337204,440 "reward_std": 0.05103103443980217,441 "rewards/accuracy_reward": 0.7291666716337204,442 "rewards/format_reward": 0.0,443 "step": 31444 },445 {446 "clip_ratio": 0.0,447 "completion_length": 345.9375114440918,448 "epoch": 0.3413333333333333,449 "grad_norm": 0.01121461670845747,450 "kl": 0.0008133649826049805,451 "learning_rate": 2.998120497826133e-06,452 "loss": 0.0072,453 "reward": 0.895833333954215,454 "reward_std": 0.05103103443980217,455 "rewards/accuracy_reward": 0.895833333954215,456 "rewards/format_reward": 0.0,457 "step": 32458 },459 {460 "clip_ratio": 0.0,461 "completion_length": 313.35417556762695,462 "epoch": 0.352,463 "grad_norm": 0.05279409885406494,464 "kl": 0.002323627471923828,465 "learning_rate": 2.9970636228970757e-06,466 "loss": -0.0081,467 "reward": 0.7916666753590107,468 "reward_std": 0.24859581515192986,469 "rewards/accuracy_reward": 0.7916666753590107,470 "rewards/format_reward": 0.0,471 "step": 33472 },473 {474 "clip_ratio": 0.0,475 "completion_length": 269.2291736602783,476 "epoch": 0.3626666666666667,477 "grad_norm": 0.054334308952093124,478 "kl": 0.002706080675125122,479 "learning_rate": 2.995772224185354e-06,480 "loss": 0.0001,481 "reward": 0.9583333358168602,482 "reward_std": 0.06454972922801971,483 "rewards/accuracy_reward": 0.9583333358168602,484 "rewards/format_reward": 0.0,485 "step": 34486 },487 {488 "clip_ratio": 0.0,489 "completion_length": 453.14584732055664,490 "epoch": 0.37333333333333335,491 "grad_norm": 0.48555248975753784,492 "kl": 0.8578243255615234,493 "learning_rate": 2.9942465039961373e-06,494 "loss": 0.1078,495 "reward": 0.687500013038516,496 "reward_std": 0.21764283254742622,497 "rewards/accuracy_reward": 0.687500013038516,498 "rewards/format_reward": 0.0,499 "step": 35500 },501 {502 "clip_ratio": 0.0,503 "completion_length": 378.81250953674316,504 "epoch": 0.384,505 "grad_norm": 0.047343045473098755,506 "kl": 0.0028142929077148438,507 "learning_rate": 2.992486701342427e-06,508 "loss": 0.0103,509 "reward": 0.7916666734963655,510 "reward_std": 0.16661179810762405,511 "rewards/accuracy_reward": 0.7916666734963655,512 "rewards/format_reward": 0.0,513 "step": 36514 },515 {516 "clip_ratio": 0.0,517 "completion_length": 407.458345413208,518 "epoch": 0.39466666666666667,519 "grad_norm": 0.37469470500946045,520 "kl": 0.9185905456542969,521 "learning_rate": 2.9904930919076145e-06,522 "loss": 0.0625,523 "reward": 0.7291666679084301,524 "reward_std": 0.1705273911356926,525 "rewards/accuracy_reward": 0.7291666679084301,526 "rewards/format_reward": 0.0,527 "step": 37528 },529 {530 "clip_ratio": 0.0,531 "completion_length": 441.22918128967285,532 "epoch": 0.4053333333333333,533 "grad_norm": 0.34893208742141724,534 "kl": 1.2932720184326172,535 "learning_rate": 2.9882659880022905e-06,536 "loss": 0.1287,537 "reward": 0.7083333376795053,538 "reward_std": 0.24859581887722015,539 "rewards/accuracy_reward": 0.7083333376795053,540 "rewards/format_reward": 0.0,541 "step": 38542 },543 {544 "clip_ratio": 0.0,545 "completion_length": 406.0208396911621,546 "epoch": 0.416,547 "grad_norm": 0.1943722516298294,548 "kl": 0.6224098205566406,549 "learning_rate": 2.985805738515323e-06,550 "loss": 0.0537,551 "reward": 0.8541666697710752,552 "reward_std": 0.11558076366782188,553 "rewards/accuracy_reward": 0.8541666697710752,554 "rewards/format_reward": 0.0,555 "step": 39556 },557 {558 "clip_ratio": 0.0,559 "completion_length": 370.4166774749756,560 "epoch": 0.4266666666666667,561 "grad_norm": 0.4532138705253601,562 "kl": 0.5034923553466797,563 "learning_rate": 2.9831127288592004e-06,564 "loss": 0.0644,565 "reward": 0.8333333358168602,566 "reward_std": 0.20148037374019623,567 "rewards/accuracy_reward": 0.8333333358168602,568 "rewards/format_reward": 0.0,569 "step": 40570 },571 {572 "clip_ratio": 0.0,573 "completion_length": 506.4166851043701,574 "epoch": 0.43733333333333335,575 "grad_norm": 0.2463473230600357,576 "kl": 1.607778549194336,577 "learning_rate": 2.9801873809096543e-06,578 "loss": 0.1184,579 "reward": 0.7291666753590107,580 "reward_std": 0.1801304928958416,581 "rewards/accuracy_reward": 0.7291666753590107,582 "rewards/format_reward": 0.0,583 "step": 41584 },585 {586 "clip_ratio": 0.0,587 "completion_length": 393.9166793823242,588 "epoch": 0.448,589 "grad_norm": 0.7764784097671509,590 "kl": 0.9653873443603516,591 "learning_rate": 2.97703015293957e-06,592 "loss": 0.1261,593 "reward": 0.8125000074505806,594 "reward_std": 0.11558076739311218,595 "rewards/accuracy_reward": 0.8125000074505806,596 "rewards/format_reward": 0.0,597 "step": 42598 },599 {600 "clip_ratio": 0.0,601 "completion_length": 422.12501525878906,602 "epoch": 0.45866666666666667,603 "grad_norm": 1.4099422693252563,604 "kl": 1.4124107360839844,605 "learning_rate": 2.9736415395471975e-06,606 "loss": 0.1981,607 "reward": 0.7708333414047956,608 "reward_std": 0.2350771203637123,609 "rewards/accuracy_reward": 0.7708333414047956,610 "rewards/format_reward": 0.0,611 "step": 43612 },613 {614 "clip_ratio": 0.0,615 "completion_length": 312.47917556762695,616 "epoch": 0.4693333333333333,617 "grad_norm": 0.00318885431624949,618 "kl": 0.0023512840270996094,619 "learning_rate": 2.970022071578666e-06,620 "loss": 0.0001,621 "reward": 1.0,622 "reward_std": 0.0,623 "rewards/accuracy_reward": 1.0,624 "rewards/format_reward": 0.0,625 "step": 44626 },627 {628 "clip_ratio": 0.0,629 "completion_length": 417.3125057220459,630 "epoch": 0.48,631 "grad_norm": 0.4240525960922241,632 "kl": 1.9218940734863281,633 "learning_rate": 2.9661723160448267e-06,634 "loss": 0.1428,635 "reward": 0.8750000055879354,636 "reward_std": 0.10206207260489464,637 "rewards/accuracy_reward": 0.8750000055879354,638 "rewards/format_reward": 0.0,639 "step": 45640 },641 {642 "clip_ratio": 0.0,643 "completion_length": 384.7083396911621,644 "epoch": 0.49066666666666664,645 "grad_norm": 0.033430058509111404,646 "kl": 0.0019054412841796875,647 "learning_rate": 2.962092876032427e-06,648 "loss": 0.0056,649 "reward": 0.8333333358168602,650 "reward_std": 0.06454972922801971,651 "rewards/accuracy_reward": 0.8333333358168602,652 "rewards/format_reward": 0.0,653 "step": 46654 },655 {656 "clip_ratio": 0.0,657 "completion_length": 459.4166793823242,658 "epoch": 0.5013333333333333,659 "grad_norm": 1.4057543277740479,660 "kl": 2.4302802085876465,661 "learning_rate": 2.957784390609632e-06,662 "loss": 0.1647,663 "reward": 0.7083333358168602,664 "reward_std": 0.06454972922801971,665 "rewards/accuracy_reward": 0.7083333358168602,666 "rewards/format_reward": 0.0,667 "step": 47668 },669 {670 "clip_ratio": 0.0,671 "completion_length": 348.0625114440918,672 "epoch": 0.512,673 "grad_norm": 0.002613195450976491,674 "kl": 0.0018477439880371094,675 "learning_rate": 2.9532475347259113e-06,676 "loss": 0.0001,677 "reward": 1.0,678 "reward_std": 0.0,679 "rewards/accuracy_reward": 1.0,680 "rewards/format_reward": 0.0,681 "step": 48682 },683 {684 "clip_ratio": 0.0,685 "completion_length": 416.31250762939453,686 "epoch": 0.5226666666666666,687 "grad_norm": 0.7081112861633301,688 "kl": 1.2155981063842773,689 "learning_rate": 2.9484830191063057e-06,690 "loss": 0.0487,691 "reward": 0.75,692 "reward_std": 0.0,693 "rewards/accuracy_reward": 0.75,694 "rewards/format_reward": 0.0,695 "step": 49696 },697 {698 "clip_ratio": 0.0,699 "completion_length": 479.18751525878906,700 "epoch": 0.5333333333333333,701 "grad_norm": 0.11860348284244537,702 "kl": 0.8176593780517578,703 "learning_rate": 2.9434915901400857e-06,704 "loss": 0.0756,705 "reward": 0.7708333414047956,706 "reward_std": 0.2446802221238613,707 "rewards/accuracy_reward": 0.7708333414047956,708 "rewards/format_reward": 0.0,709 "step": 50710 },711 {712 "clip_ratio": 0.0,713 "completion_length": 336.14584159851074,714 "epoch": 0.544,715 "grad_norm": 0.05025426298379898,716 "kl": 0.0013027191162109375,717 "learning_rate": 2.938274029763826e-06,718 "loss": 0.0322,719 "reward": 0.8125000074505806,720 "reward_std": 0.11558076739311218,721 "rewards/accuracy_reward": 0.8125000074505806,722 "rewards/format_reward": 0.0,723 "step": 51724 },725 {726 "clip_ratio": 0.0,727 "completion_length": 414.06250953674316,728 "epoch": 0.5546666666666666,729 "grad_norm": 0.5244683623313904,730 "kl": 1.343430519104004,731 "learning_rate": 2.9328311553389116e-06,732 "loss": 0.091,733 "reward": 0.791666679084301,734 "reward_std": 0.16661180555820465,735 "rewards/accuracy_reward": 0.791666679084301,736 "rewards/format_reward": 0.0,737 "step": 52738 },739 {740 "clip_ratio": 0.0,741 "completion_length": 418.7083435058594,742 "epoch": 0.5653333333333334,743 "grad_norm": 0.29313525557518005,744 "kl": 0.7863144874572754,745 "learning_rate": 2.927163819523492e-06,746 "loss": 0.0602,747 "reward": 0.6041666679084301,748 "reward_std": 0.1705273911356926,749 "rewards/accuracy_reward": 0.6041666679084301,750 "rewards/format_reward": 0.0,751 "step": 53752 },753 {754 "clip_ratio": 0.0,755 "completion_length": 369.54168128967285,756 "epoch": 0.576,757 "grad_norm": 0.413883239030838,758 "kl": 0.3720831871032715,759 "learning_rate": 2.9212729101389084e-06,760 "loss": 0.0502,761 "reward": 0.7916666734963655,762 "reward_std": 0.16661179810762405,763 "rewards/accuracy_reward": 0.7916666734963655,764 "rewards/format_reward": 0.0,765 "step": 54766 },767 {768 "clip_ratio": 0.0,769 "completion_length": 555.5625190734863,770 "epoch": 0.5866666666666667,771 "grad_norm": 0.8179274797439575,772 "kl": 1.2964344024658203,773 "learning_rate": 2.9151593500306103e-06,774 "loss": 0.0543,775 "reward": 0.5625,776 "reward_std": 0.06846532225608826,777 "rewards/accuracy_reward": 0.5625,778 "rewards/format_reward": 0.0,779 "step": 55780 },781 {782 "clip_ratio": 0.0,783 "completion_length": 374.7291736602783,784 "epoch": 0.5973333333333334,785 "grad_norm": 0.002118342788890004,786 "kl": 0.0013017654418945312,787 "learning_rate": 2.9088240969235863e-06,788 "loss": 0.0001,789 "reward": 1.0,790 "reward_std": 0.0,791 "rewards/accuracy_reward": 1.0,792 "rewards/format_reward": 0.0,793 "step": 56794 },795 {796 "clip_ratio": 0.0,797 "completion_length": 328.9166793823242,798 "epoch": 0.608,799 "grad_norm": 0.0338607020676136,800 "kl": 0.0011858940124511719,801 "learning_rate": 2.902268143272333e-06,802 "loss": -0.0089,803 "reward": 0.9583333358168602,804 "reward_std": 0.06454972922801971,805 "rewards/accuracy_reward": 0.9583333358168602,806 "rewards/format_reward": 0.0,807 "step": 57808 },809 {810 "clip_ratio": 0.0,811 "completion_length": 351.5416793823242,812 "epoch": 0.6186666666666667,813 "grad_norm": 0.5058808922767639,814 "kl": 0.20941162109375,815 "learning_rate": 2.895492516105378e-06,816 "loss": 0.0457,817 "reward": 0.8125000074505806,818 "reward_std": 0.11558076366782188,819 "rewards/accuracy_reward": 0.8125000074505806,820 "rewards/format_reward": 0.0,821 "step": 58822 },823 {824 "clip_ratio": 0.0,825 "completion_length": 574.2708396911621,826 "epoch": 0.6293333333333333,827 "grad_norm": 0.6014701128005981,828 "kl": 0.8367283344268799,829 "learning_rate": 2.888498276864393e-06,830 "loss": 0.0236,831 "reward": 0.6458333358168602,832 "reward_std": 0.13301505148410797,833 "rewards/accuracy_reward": 0.6458333358168602,834 "rewards/format_reward": 0.0,835 "step": 59836 },837 {838 "clip_ratio": 0.0,839 "completion_length": 444.9166774749756,840 "epoch": 0.64,841 "grad_norm": 0.07577009499073029,842 "kl": 0.07219600677490234,843 "learning_rate": 2.8812865212379087e-06,844 "loss": 0.0071,845 "reward": 0.8125,846 "reward_std": 0.06846532225608826,847 "rewards/accuracy_reward": 0.8125,848 "rewards/format_reward": 0.0,849 "step": 60850 },851 {852 "clip_ratio": 0.0,853 "completion_length": 378.2291736602783,854 "epoch": 0.6506666666666666,855 "grad_norm": 0.0443667508661747,856 "kl": 0.0021257400512695312,857 "learning_rate": 2.8738583789896743e-06,858 "loss": 0.0073,859 "reward": 0.9583333432674408,860 "reward_std": 0.10206207633018494,861 "rewards/accuracy_reward": 0.9583333432674408,862 "rewards/format_reward": 0.0,863 "step": 61864 },865 {866 "clip_ratio": 0.0,867 "completion_length": 514.6458396911621,868 "epoch": 0.6613333333333333,869 "grad_norm": 0.8298602104187012,870 "kl": 0.23528003692626953,871 "learning_rate": 2.866215013781668e-06,872 "loss": 0.0501,873 "reward": 0.7291666753590107,874 "reward_std": 0.1801304966211319,875 "rewards/accuracy_reward": 0.7291666753590107,876 "rewards/format_reward": 0.0,877 "step": 62878 },879 {880 "clip_ratio": 0.0,881 "completion_length": 423.06250762939453,882 "epoch": 0.672,883 "grad_norm": 0.5626036524772644,884 "kl": 0.08101940155029297,885 "learning_rate": 2.858357622991806e-06,886 "loss": 0.0213,887 "reward": 0.9583333432674408,888 "reward_std": 0.10206207260489464,889 "rewards/accuracy_reward": 0.9583333432674408,890 "rewards/format_reward": 0.0,891 "step": 63892 },893 {894 "clip_ratio": 0.0,895 "completion_length": 396.2916774749756,896 "epoch": 0.6826666666666666,897 "grad_norm": 0.4362681210041046,898 "kl": 0.0890655517578125,899 "learning_rate": 2.8502874375263646e-06,900 "loss": 0.0404,901 "reward": 0.8958333432674408,902 "reward_std": 0.1801304966211319,903 "rewards/accuracy_reward": 0.8958333432674408,904 "rewards/format_reward": 0.0,905 "step": 64906 },907 {908 "clip_ratio": 0.0,909 "completion_length": 349.8958435058594,910 "epoch": 0.6933333333333334,911 "grad_norm": 0.42613545060157776,912 "kl": 0.058725833892822266,913 "learning_rate": 2.8420057216271533e-06,914 "loss": 0.0178,915 "reward": 0.9583333432674408,916 "reward_std": 0.10206206887960434,917 "rewards/accuracy_reward": 0.9583333432674408,918 "rewards/format_reward": 0.0,919 "step": 65920 },921 {922 "clip_ratio": 0.0,923 "completion_length": 425.60417556762695,924 "epoch": 0.704,925 "grad_norm": 0.44066762924194336,926 "kl": 0.1816251277923584,927 "learning_rate": 2.8335137726734607e-06,928 "loss": 0.0316,929 "reward": 0.9583333358168602,930 "reward_std": 0.06454972922801971,931 "rewards/accuracy_reward": 0.9583333358168602,932 "rewards/format_reward": 0.0,933 "step": 66934 },935 {936 "clip_ratio": 0.0,937 "completion_length": 369.0833435058594,938 "epoch": 0.7146666666666667,939 "grad_norm": 0.12847420573234558,940 "kl": 0.4885525703430176,941 "learning_rate": 2.8248129209788164e-06,942 "loss": 0.061,943 "reward": 0.8541666753590107,944 "reward_std": 0.1801304928958416,945 "rewards/accuracy_reward": 0.8541666753590107,946 "rewards/format_reward": 0.0,947 "step": 67948 },949 {950 "clip_ratio": 0.0,951 "completion_length": 384.7708435058594,952 "epoch": 0.7253333333333334,953 "grad_norm": 0.29157620668411255,954 "kl": 0.36469125747680664,955 "learning_rate": 2.8159045295825857e-06,956 "loss": 0.0067,957 "reward": 0.7916666716337204,958 "reward_std": 0.12909945845603943,959 "rewards/accuracy_reward": 0.7916666716337204,960 "rewards/format_reward": 0.0,961 "step": 68962 },963 {964 "clip_ratio": 0.0,965 "completion_length": 307.75000953674316,966 "epoch": 0.736,967 "grad_norm": 0.02736111357808113,968 "kl": 0.001550436019897461,969 "learning_rate": 2.806789994036445e-06,970 "loss": -0.0033,971 "reward": 0.9791666716337204,972 "reward_std": 0.05103103443980217,973 "rewards/accuracy_reward": 0.9791666716337204,974 "rewards/format_reward": 0.0,975 "step": 69976 },977 {978 "clip_ratio": 0.0,979 "completion_length": 382.1250057220459,980 "epoch": 0.7466666666666667,981 "grad_norm": 0.057907938957214355,982 "kl": 0.44846928119659424,983 "learning_rate": 2.7974707421857584e-06,984 "loss": 0.0653,985 "reward": 0.833333333954215,986 "reward_std": 0.11949636042118073,987 "rewards/accuracy_reward": 0.833333333954215,988 "rewards/format_reward": 0.0,989 "step": 70990 },991 {992 "clip_ratio": 0.0,993 "completion_length": 505.02084732055664,994 "epoch": 0.7573333333333333,995 "grad_norm": 0.9753615856170654,996 "kl": 1.7791624069213867,997 "learning_rate": 2.7879482339458973e-06,998 "loss": 0.1034,999 "reward": 0.708333333954215,1000 "reward_std": 0.11949635669589043,1001 "rewards/accuracy_reward": 0.708333333954215,1002 "rewards/format_reward": 0.0,1003 "step": 711004 },1005 {1006 "clip_ratio": 0.0,1007 "completion_length": 432.9791831970215,1008 "epoch": 0.768,1009 "grad_norm": 0.3005034625530243,1010 "kl": 0.9472742080688477,1011 "learning_rate": 2.778223961073537e-06,1012 "loss": 0.0849,1013 "reward": 0.6666666716337204,1014 "reward_std": 0.12909945845603943,1015 "rewards/accuracy_reward": 0.6666666716337204,1016 "rewards/format_reward": 0.0,1017 "step": 721018 },1019 {1020 "clip_ratio": 0.0,1021 "completion_length": 353.5416793823242,1022 "epoch": 0.7786666666666666,1023 "grad_norm": 0.2883008122444153,1024 "kl": 0.44875526428222656,1025 "learning_rate": 2.7682994469329646e-06,1026 "loss": 0.0592,1027 "reward": 0.8750000074505806,1028 "reward_std": 0.18404608592391014,1029 "rewards/accuracy_reward": 0.8750000074505806,1030 "rewards/format_reward": 0.0,1031 "step": 731032 },1033 {1034 "clip_ratio": 0.0,1035 "completion_length": 426.81251525878906,1036 "epoch": 0.7893333333333333,1037 "grad_norm": 0.286614328622818,1038 "kl": 0.5456581115722656,1039 "learning_rate": 2.7581762462574334e-06,1040 "loss": 0.0588,1041 "reward": 0.8125000074505806,1042 "reward_std": 0.2621145099401474,1043 "rewards/accuracy_reward": 0.8125000074505806,1044 "rewards/format_reward": 0.0,1045 "step": 741046 },1047 {1048 "clip_ratio": 0.0,1049 "completion_length": 589.6875152587891,1050 "epoch": 0.8,1051 "grad_norm": 0.09037178009748459,1052 "kl": 1.1991617679595947,1053 "learning_rate": 2.747855944905606e-06,1054 "loss": 0.0889,1055 "reward": 0.6666666772216558,1056 "reward_std": 0.22155842930078506,1057 "rewards/accuracy_reward": 0.6666666772216558,1058 "rewards/format_reward": 0.0,1059 "step": 751060 },1061 {1062 "clip_ratio": 0.0,1063 "completion_length": 396.75001525878906,1064 "epoch": 0.8106666666666666,1065 "grad_norm": 0.04595573991537094,1066 "kl": 0.0017056465148925781,1067 "learning_rate": 2.73734015961312e-06,1068 "loss": 0.0054,1069 "reward": 0.8958333432674408,1070 "reward_std": 0.1705273911356926,1071 "rewards/accuracy_reward": 0.8958333432674408,1072 "rewards/format_reward": 0.0,1073 "step": 761074 },1075 {1076 "clip_ratio": 0.0,1077 "completion_length": 421.25001525878906,1078 "epoch": 0.8213333333333334,1079 "grad_norm": 0.6178137063980103,1080 "kl": 0.7442665100097656,1081 "learning_rate": 2.7266305377393183e-06,1082 "loss": 0.1024,1083 "reward": 0.8541666716337204,1084 "reward_std": 0.18796168267726898,1085 "rewards/accuracy_reward": 0.8541666716337204,1086 "rewards/format_reward": 0.0,1087 "step": 771088 },1089 {1090 "clip_ratio": 0.0,1091 "completion_length": 423.0625114440918,1092 "epoch": 0.832,1093 "grad_norm": 0.040635429322719574,1094 "kl": 0.18666815757751465,1095 "learning_rate": 2.7157287570091765e-06,1096 "loss": 0.0134,1097 "reward": 0.8541666679084301,1098 "reward_std": 0.13301505148410797,1099 "rewards/accuracy_reward": 0.8541666679084301,1100 "rewards/format_reward": 0.0,1101 "step": 781102 },1103 {1104 "clip_ratio": 0.0,1105 "completion_length": 460.87500762939453,1106 "epoch": 0.8426666666666667,1107 "grad_norm": 0.2019713968038559,1108 "kl": 0.32639598846435547,1109 "learning_rate": 2.7046365252504835e-06,1110 "loss": 0.029,1111 "reward": 0.8333333432674408,1112 "reward_std": 0.10206206887960434,1113 "rewards/accuracy_reward": 0.8333333432674408,1114 "rewards/format_reward": 0.0,1115 "step": 791116 },1117 {1118 "clip_ratio": 0.0,1119 "completion_length": 368.5000114440918,1120 "epoch": 0.8533333333333334,1121 "grad_norm": 0.07597187906503677,1122 "kl": 1.0420165061950684,1123 "learning_rate": 2.693355580126295e-06,1124 "loss": 0.0706,1125 "reward": 0.8541666679084301,1126 "reward_std": 0.13301505148410797,1127 "rewards/accuracy_reward": 0.8541666679084301,1128 "rewards/format_reward": 0.0,1129 "step": 801130 },1131 {1132 "clip_ratio": 0.0,1133 "completion_length": 421.4375057220459,1134 "epoch": 0.864,1135 "grad_norm": 0.33045217394828796,1136 "kl": 0.7187583446502686,1137 "learning_rate": 2.6818876888627203e-06,1138 "loss": 0.0402,1139 "reward": 0.7500000074505806,1140 "reward_std": 0.18404608592391014,1141 "rewards/accuracy_reward": 0.7500000074505806,1142 "rewards/format_reward": 0.0,1143 "step": 811144 },1145 {1146 "clip_ratio": 0.0,1147 "completion_length": 438.87500762939453,1148 "epoch": 0.8746666666666667,1149 "grad_norm": 1.0251787900924683,1150 "kl": 2.417191505432129,1151 "learning_rate": 2.6702346479720777e-06,1152 "loss": 0.132,1153 "reward": 0.7291666716337204,1154 "reward_std": 0.05103103443980217,1155 "rewards/accuracy_reward": 0.7291666716337204,1156 "rewards/format_reward": 0.0,1157 "step": 821158 },1159 {1160 "clip_ratio": 0.0,1161 "completion_length": 351.25000762939453,1162 "epoch": 0.8853333333333333,1163 "grad_norm": 0.25971361994743347,1164 "kl": 0.2927316427230835,1165 "learning_rate": 2.6583982829714576e-06,1166 "loss": 0.0172,1167 "reward": 0.875,1168 "reward_std": 0.0,1169 "rewards/accuracy_reward": 0.875,1170 "rewards/format_reward": 0.0,1171 "step": 831172 },1173 {1174 "clip_ratio": 0.0,1175 "completion_length": 382.8541736602783,1176 "epoch": 0.896,1177 "grad_norm": 0.3502584993839264,1178 "kl": 0.24036216735839844,1179 "learning_rate": 2.6463804480967447e-06,1180 "loss": 0.0407,1181 "reward": 0.9375000074505806,1182 "reward_std": 0.11558076366782188,1183 "rewards/accuracy_reward": 0.9375000074505806,1184 "rewards/format_reward": 0.0,1185 "step": 841186 },1187 {1188 "clip_ratio": 0.0,1189 "completion_length": 392.93751335144043,1190 "epoch": 0.9066666666666666,1191 "grad_norm": 0.3297070562839508,1192 "kl": 0.43697142601013184,1193 "learning_rate": 2.6341830260121445e-06,1194 "loss": 0.0187,1195 "reward": 0.7083333358168602,1196 "reward_std": 0.06454972922801971,1197 "rewards/accuracy_reward": 0.7083333358168602,1198 "rewards/format_reward": 0.0,1199 "step": 851200 },