sydqiang/X-R1-0.5B
011
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": 574.645845413208,14 "epoch": 0.010666666666666666,15 "grad_norm": 0.08630427718162537,16 "kl": 0.0,17 "learning_rate": 1.0714285714285714e-07,18 "loss": -0.0341,19 "reward": 0.06250000186264515,20 "reward_std": 0.11558076366782188,21 "rewards/accuracy_reward": 0.06250000186264515,22 "rewards/format_reward": 0.0,23 "step": 124 },25 {26 "clip_ratio": 0.0,27 "completion_length": 611.5833511352539,28 "epoch": 0.021333333333333333,29 "grad_norm": 0.06199095398187637,30 "kl": 0.0,31 "learning_rate": 2.1428571428571428e-07,32 "loss": -0.0023,33 "reward": 0.02083333395421505,34 "reward_std": 0.05103103443980217,35 "rewards/accuracy_reward": 0.02083333395421505,36 "rewards/format_reward": 0.0,37 "step": 238 },39 {40 "clip_ratio": 0.0,41 "completion_length": 514.1250114440918,42 "epoch": 0.032,43 "grad_norm": 0.04586375877261162,44 "kl": 0.000579833984375,45 "learning_rate": 3.214285714285714e-07,46 "loss": 0.0497,47 "reward": 0.02083333395421505,48 "reward_std": 0.05103103443980217,49 "rewards/accuracy_reward": 0.02083333395421505,50 "rewards/format_reward": 0.0,51 "step": 352 },53 {54 "clip_ratio": 0.0,55 "completion_length": 554.4375152587891,56 "epoch": 0.042666666666666665,57 "grad_norm": 0.0028707189485430717,58 "kl": 0.0006541013717651367,59 "learning_rate": 4.2857142857142857e-07,60 "loss": 0.0,61 "reward": 0.0,62 "reward_std": 0.0,63 "rewards/accuracy_reward": 0.0,64 "rewards/format_reward": 0.0,65 "step": 466 },67 {68 "clip_ratio": 0.0,69 "completion_length": 508.04168701171875,70 "epoch": 0.05333333333333334,71 "grad_norm": 0.1056743636727333,72 "kl": 0.0003535747528076172,73 "learning_rate": 5.357142857142858e-07,74 "loss": 0.0614,75 "reward": 0.06250000186264515,76 "reward_std": 0.11558076739311218,77 "rewards/accuracy_reward": 0.06250000186264515,78 "rewards/format_reward": 0.0,79 "step": 580 },81 {82 "clip_ratio": 0.0,83 "completion_length": 543.0833511352539,84 "epoch": 0.064,85 "grad_norm": 0.04604552313685417,86 "kl": 0.00043952465057373047,87 "learning_rate": 6.428571428571428e-07,88 "loss": -0.0038,89 "reward": 0.02083333395421505,90 "reward_std": 0.05103103816509247,91 "rewards/accuracy_reward": 0.02083333395421505,92 "rewards/format_reward": 0.0,93 "step": 694 },95 {96 "clip_ratio": 0.0,97 "completion_length": 585.8958511352539,98 "epoch": 0.07466666666666667,99 "grad_norm": 0.06841025501489639,100 "kl": 0.00035643577575683594,101 "learning_rate": 7.5e-07,102 "loss": 0.0413,103 "reward": 0.0416666679084301,104 "reward_std": 0.06454972922801971,105 "rewards/accuracy_reward": 0.0416666679084301,106 "rewards/format_reward": 0.0,107 "step": 7108 },109 {110 "clip_ratio": 0.0,111 "completion_length": 563.3541870117188,112 "epoch": 0.08533333333333333,113 "grad_norm": 0.0019720818381756544,114 "kl": 0.0005422830581665039,115 "learning_rate": 8.571428571428571e-07,116 "loss": 0.0,117 "reward": 0.0,118 "reward_std": 0.0,119 "rewards/accuracy_reward": 0.0,120 "rewards/format_reward": 0.0,121 "step": 8122 },123 {124 "clip_ratio": 0.0,125 "completion_length": 643.0625152587891,126 "epoch": 0.096,127 "grad_norm": 0.0895163044333458,128 "kl": 0.00038993358612060547,129 "learning_rate": 9.642857142857145e-07,130 "loss": -0.044,131 "reward": 0.02083333395421505,132 "reward_std": 0.05103103443980217,133 "rewards/accuracy_reward": 0.02083333395421505,134 "rewards/format_reward": 0.0,135 "step": 9136 },137 {138 "clip_ratio": 0.0,139 "completion_length": 607.1458511352539,140 "epoch": 0.10666666666666667,141 "grad_norm": 0.03434349223971367,142 "kl": 0.00034165382385253906,143 "learning_rate": 1.0714285714285716e-06,144 "loss": 0.0228,145 "reward": 0.02083333395421505,146 "reward_std": 0.05103103443980217,147 "rewards/accuracy_reward": 0.02083333395421505,148 "rewards/format_reward": 0.0,149 "step": 10150 },151 {152 "clip_ratio": 0.0,153 "completion_length": 528.0000152587891,154 "epoch": 0.11733333333333333,155 "grad_norm": 0.2434433549642563,156 "kl": 0.0005402565002441406,157 "learning_rate": 1.1785714285714285e-06,158 "loss": -0.0169,159 "reward": 0.0416666679084301,160 "reward_std": 0.10206207633018494,161 "rewards/accuracy_reward": 0.0416666679084301,162 "rewards/format_reward": 0.0,163 "step": 11164 },165 {166 "clip_ratio": 0.0,167 "completion_length": 483.83335876464844,168 "epoch": 0.128,169 "grad_norm": 0.002469642786309123,170 "kl": 0.0004532337188720703,171 "learning_rate": 1.2857142857142856e-06,172 "loss": 0.0,173 "reward": 0.0,174 "reward_std": 0.0,175 "rewards/accuracy_reward": 0.0,176 "rewards/format_reward": 0.0,177 "step": 12178 },179 {180 "clip_ratio": 0.0,181 "completion_length": 513.5625038146973,182 "epoch": 0.13866666666666666,183 "grad_norm": 0.14749591052532196,184 "kl": 0.000500798225402832,185 "learning_rate": 1.392857142857143e-06,186 "loss": -0.0378,187 "reward": 0.0416666679084301,188 "reward_std": 0.10206207633018494,189 "rewards/accuracy_reward": 0.0416666679084301,190 "rewards/format_reward": 0.0,191 "step": 13192 },193 {194 "clip_ratio": 0.0,195 "completion_length": 538.7916831970215,196 "epoch": 0.14933333333333335,197 "grad_norm": 0.161040261387825,198 "kl": 0.0008957386016845703,199 "learning_rate": 1.5e-06,200 "loss": 0.0062,201 "reward": 0.0416666679084301,202 "reward_std": 0.10206206887960434,203 "rewards/accuracy_reward": 0.0416666679084301,204 "rewards/format_reward": 0.0,205 "step": 14206 },207 {208 "clip_ratio": 0.0,209 "completion_length": 596.1875114440918,210 "epoch": 0.16,211 "grad_norm": 0.0028237502556294203,212 "kl": 0.0005574226379394531,213 "learning_rate": 1.6071428571428572e-06,214 "loss": 0.0,215 "reward": 0.0,216 "reward_std": 0.0,217 "rewards/accuracy_reward": 0.0,218 "rewards/format_reward": 0.0,219 "step": 15220 },221 {222 "clip_ratio": 0.0,223 "completion_length": 534.1458587646484,224 "epoch": 0.17066666666666666,225 "grad_norm": 0.25467991828918457,226 "kl": 0.0010123252868652344,227 "learning_rate": 1.7142857142857143e-06,228 "loss": -0.0508,229 "reward": 0.0416666679084301,230 "reward_std": 0.10206207260489464,231 "rewards/accuracy_reward": 0.0416666679084301,232 "rewards/format_reward": 0.0,233 "step": 16234 },235 {236 "clip_ratio": 0.0,237 "completion_length": 484.0416851043701,238 "epoch": 0.18133333333333335,239 "grad_norm": 0.011162322014570236,240 "kl": 0.0026879310607910156,241 "learning_rate": 1.8214285714285714e-06,242 "loss": 0.0,243 "reward": 0.0,244 "reward_std": 0.0,245 "rewards/accuracy_reward": 0.0,246 "rewards/format_reward": 0.0,247 "step": 17248 },249 {250 "clip_ratio": 0.0,251 "completion_length": 616.6875114440918,252 "epoch": 0.192,253 "grad_norm": 0.10696219652891159,254 "kl": 0.0013022422790527344,255 "learning_rate": 1.928571428571429e-06,256 "loss": 0.0673,257 "reward": 0.06250000186264515,258 "reward_std": 0.1530931033194065,259 "rewards/accuracy_reward": 0.06250000186264515,260 "rewards/format_reward": 0.0,261 "step": 18262 },263 {264 "clip_ratio": 0.0,265 "completion_length": 542.3750152587891,266 "epoch": 0.20266666666666666,267 "grad_norm": 0.10040425509214401,268 "kl": 0.001857757568359375,269 "learning_rate": 2.035714285714286e-06,270 "loss": 0.0068,271 "reward": 0.02083333395421505,272 "reward_std": 0.05103103816509247,273 "rewards/accuracy_reward": 0.02083333395421505,274 "rewards/format_reward": 0.0,275 "step": 19276 },277 {278 "clip_ratio": 0.0,279 "completion_length": 631.1041831970215,280 "epoch": 0.21333333333333335,281 "grad_norm": 0.07833883166313171,282 "kl": 0.002624988555908203,283 "learning_rate": 2.142857142857143e-06,284 "loss": 0.0135,285 "reward": 0.0416666679084301,286 "reward_std": 0.10206207260489464,287 "rewards/accuracy_reward": 0.0416666679084301,288 "rewards/format_reward": 0.0,289 "step": 20290 },291 {292 "clip_ratio": 0.0,293 "completion_length": 422.77084159851074,294 "epoch": 0.224,295 "grad_norm": 0.11672168970108032,296 "kl": 0.0066432952880859375,297 "learning_rate": 2.25e-06,298 "loss": -0.0038,299 "reward": 0.0625,300 "reward_std": 0.06846532225608826,301 "rewards/accuracy_reward": 0.0625,302 "rewards/format_reward": 0.0,303 "step": 21304 },305 {306 "clip_ratio": 0.0,307 "completion_length": 511.95834732055664,308 "epoch": 0.23466666666666666,309 "grad_norm": 0.05832681804895401,310 "kl": 0.004792213439941406,311 "learning_rate": 2.357142857142857e-06,312 "loss": 0.0088,313 "reward": 0.0625,314 "reward_std": 0.06846532225608826,315 "rewards/accuracy_reward": 0.0625,316 "rewards/format_reward": 0.0,317 "step": 22318 },319 {320 "clip_ratio": 0.0,321 "completion_length": 454.6666793823242,322 "epoch": 0.24533333333333332,323 "grad_norm": 0.05373668298125267,324 "kl": 0.006567955017089844,325 "learning_rate": 2.4642857142857143e-06,326 "loss": 0.0188,327 "reward": 0.02083333395421505,328 "reward_std": 0.05103103443980217,329 "rewards/accuracy_reward": 0.02083333395421505,330 "rewards/format_reward": 0.0,331 "step": 23332 },333 {334 "clip_ratio": 0.0,335 "completion_length": 457.27084732055664,336 "epoch": 0.256,337 "grad_norm": 0.18046098947525024,338 "kl": 0.0054378509521484375,339 "learning_rate": 2.571428571428571e-06,340 "loss": 0.011,341 "reward": 0.06250000186264515,342 "reward_std": 0.11558076366782188,343 "rewards/accuracy_reward": 0.06250000186264515,344 "rewards/format_reward": 0.0,345 "step": 24346 },347 {348 "clip_ratio": 0.0,349 "completion_length": 514.3333473205566,350 "epoch": 0.26666666666666666,351 "grad_norm": 0.1545768529176712,352 "kl": 0.006916046142578125,353 "learning_rate": 2.678571428571429e-06,354 "loss": 0.0313,355 "reward": 0.0416666679084301,356 "reward_std": 0.10206207260489464,357 "rewards/accuracy_reward": 0.0416666679084301,358 "rewards/format_reward": 0.0,359 "step": 25360 },361 {362 "clip_ratio": 0.0,363 "completion_length": 543.0833511352539,364 "epoch": 0.2773333333333333,365 "grad_norm": 0.08341816812753677,366 "kl": 0.0072002410888671875,367 "learning_rate": 2.785714285714286e-06,368 "loss": -0.0054,369 "reward": 0.02083333395421505,370 "reward_std": 0.05103103443980217,371 "rewards/accuracy_reward": 0.02083333395421505,372 "rewards/format_reward": 0.0,373 "step": 26374 },375 {376 "clip_ratio": 0.0,377 "completion_length": 500.02084732055664,378 "epoch": 0.288,379 "grad_norm": 0.1588888019323349,380 "kl": 0.018650054931640625,381 "learning_rate": 2.892857142857143e-06,382 "loss": -0.0055,383 "reward": 0.02083333395421505,384 "reward_std": 0.05103103443980217,385 "rewards/accuracy_reward": 0.02083333395421505,386 "rewards/format_reward": 0.0,387 "step": 27388 },389 {390 "clip_ratio": 0.0,391 "completion_length": 517.6041889190674,392 "epoch": 0.2986666666666667,393 "grad_norm": 0.1424867957830429,394 "kl": 0.007252693176269531,395 "learning_rate": 3e-06,396 "loss": -0.003,397 "reward": 0.0416666679084301,398 "reward_std": 0.10206206887960434,399 "rewards/accuracy_reward": 0.0416666679084301,400 "rewards/format_reward": 0.0,401 "step": 28402 },403 {404 "clip_ratio": 0.0,405 "completion_length": 477.33334732055664,406 "epoch": 0.30933333333333335,407 "grad_norm": 0.11833926290273666,408 "kl": 0.01897430419921875,409 "learning_rate": 2.999882508108334e-06,410 "loss": 0.0413,411 "reward": 0.08333333395421505,412 "reward_std": 0.11949636042118073,413 "rewards/accuracy_reward": 0.08333333395421505,414 "rewards/format_reward": 0.0,415 "step": 29416 },417 {418 "clip_ratio": 0.0,419 "completion_length": 529.1041851043701,420 "epoch": 0.32,421 "grad_norm": 0.21591709554195404,422 "kl": 0.010650634765625,423 "learning_rate": 2.9995300508391286e-06,424 "loss": -0.0722,425 "reward": 0.16666667349636555,426 "reward_std": 0.20412414520978928,427 "rewards/accuracy_reward": 0.16666667349636555,428 "rewards/format_reward": 0.0,429 "step": 30430 },431 {432 "clip_ratio": 0.0,433 "completion_length": 487.5416831970215,434 "epoch": 0.33066666666666666,435 "grad_norm": 0.12658198177814484,436 "kl": 0.011867523193359375,437 "learning_rate": 2.998942683406879e-06,438 "loss": 0.0104,439 "reward": 0.18750000558793545,440 "reward_std": 0.2446802221238613,441 "rewards/accuracy_reward": 0.18750000558793545,442 "rewards/format_reward": 0.0,443 "step": 31444 },445 {446 "clip_ratio": 0.0,447 "completion_length": 460.0625114440918,448 "epoch": 0.3413333333333333,449 "grad_norm": 0.21254189312458038,450 "kl": 0.02234649658203125,451 "learning_rate": 2.998120497826133e-06,452 "loss": 0.0341,453 "reward": 0.08333333395421505,454 "reward_std": 0.11949635669589043,455 "rewards/accuracy_reward": 0.08333333395421505,456 "rewards/format_reward": 0.0,457 "step": 32458 },459 {460 "clip_ratio": 0.0,461 "completion_length": 501.5625114440918,462 "epoch": 0.352,463 "grad_norm": 0.18026530742645264,464 "kl": 0.013641357421875,465 "learning_rate": 2.9970636228970757e-06,466 "loss": -0.0473,467 "reward": 0.12500000186264515,468 "reward_std": 0.22155842930078506,469 "rewards/accuracy_reward": 0.12500000186264515,470 "rewards/format_reward": 0.0,471 "step": 33472 },473 {474 "clip_ratio": 0.0,475 "completion_length": 513.2916774749756,476 "epoch": 0.3626666666666667,477 "grad_norm": 0.1767341047525406,478 "kl": 0.021457672119140625,479 "learning_rate": 2.995772224185354e-06,480 "loss": 0.0094,481 "reward": 0.0833333358168602,482 "reward_std": 0.16661180183291435,483 "rewards/accuracy_reward": 0.0833333358168602,484 "rewards/format_reward": 0.0,485 "step": 34486 },487 {488 "clip_ratio": 0.0,489 "completion_length": 660.7083511352539,490 "epoch": 0.37333333333333335,491 "grad_norm": 0.13715602457523346,492 "kl": 0.022609710693359375,493 "learning_rate": 2.9942465039961373e-06,494 "loss": -0.0303,495 "reward": 0.06250000186264515,496 "reward_std": 0.1530931070446968,497 "rewards/accuracy_reward": 0.06250000186264515,498 "rewards/format_reward": 0.0,499 "step": 35500 },501 {502 "clip_ratio": 0.0,503 "completion_length": 580.8125152587891,504 "epoch": 0.384,505 "grad_norm": 0.11364670842885971,506 "kl": 0.011260986328125,507 "learning_rate": 2.992486701342427e-06,508 "loss": 0.0458,509 "reward": 0.1458333358168602,510 "reward_std": 0.2350771278142929,511 "rewards/accuracy_reward": 0.1458333358168602,512 "rewards/format_reward": 0.0,513 "step": 36514 },515 {516 "clip_ratio": 0.0,517 "completion_length": 500.08334732055664,518 "epoch": 0.39466666666666667,519 "grad_norm": 0.09874270111322403,520 "kl": 0.015888214111328125,521 "learning_rate": 2.9904930919076145e-06,522 "loss": -0.0039,523 "reward": 0.0833333358168602,524 "reward_std": 0.06454972922801971,525 "rewards/accuracy_reward": 0.0833333358168602,526 "rewards/format_reward": 0.0,527 "step": 37528 },529 {530 "clip_ratio": 0.0,531 "completion_length": 628.7708511352539,532 "epoch": 0.4053333333333333,533 "grad_norm": 0.089125856757164,534 "kl": 0.011135101318359375,535 "learning_rate": 2.9882659880022905e-06,536 "loss": -0.0547,537 "reward": 0.12500000186264515,538 "reward_std": 0.18404608964920044,539 "rewards/accuracy_reward": 0.12500000186264515,540 "rewards/format_reward": 0.0,541 "step": 38542 },543 {544 "clip_ratio": 0.0,545 "completion_length": 614.1666889190674,546 "epoch": 0.416,547 "grad_norm": 0.2361649125814438,548 "kl": 0.016826629638671875,549 "learning_rate": 2.985805738515323e-06,550 "loss": 0.0092,551 "reward": 0.18750000558793545,552 "reward_std": 0.1801304928958416,553 "rewards/accuracy_reward": 0.18750000558793545,554 "rewards/format_reward": 0.0,555 "step": 39556 },557 {558 "clip_ratio": 0.0,559 "completion_length": 500.7708492279053,560 "epoch": 0.4266666666666667,561 "grad_norm": 0.2772132158279419,562 "kl": 0.01801300048828125,563 "learning_rate": 2.9831127288592004e-06,564 "loss": -0.0077,565 "reward": 0.2291666753590107,566 "reward_std": 0.28219256550073624,567 "rewards/accuracy_reward": 0.2291666753590107,568 "rewards/format_reward": 0.0,569 "step": 40570 },571 {572 "clip_ratio": 0.0,573 "completion_length": 615.9375114440918,574 "epoch": 0.43733333333333335,575 "grad_norm": 0.11632812768220901,576 "kl": 0.013370513916015625,577 "learning_rate": 2.9801873809096543e-06,578 "loss": 0.0736,579 "reward": 0.18750000558793545,580 "reward_std": 0.2446802258491516,581 "rewards/accuracy_reward": 0.18750000558793545,582 "rewards/format_reward": 0.0,583 "step": 41584 },585 {586 "clip_ratio": 0.0,587 "completion_length": 481.37500762939453,588 "epoch": 0.448,589 "grad_norm": 0.16239772737026215,590 "kl": 0.02106475830078125,591 "learning_rate": 2.97703015293957e-06,592 "loss": 0.018,593 "reward": 0.20833333767950535,594 "reward_std": 0.3506578877568245,595 "rewards/accuracy_reward": 0.20833333767950535,596 "rewards/format_reward": 0.0,597 "step": 42598 },599 {600 "clip_ratio": 0.0,601 "completion_length": 447.4166831970215,602 "epoch": 0.45866666666666667,603 "grad_norm": 0.17852915823459625,604 "kl": 0.027374267578125,605 "learning_rate": 2.9736415395471975e-06,606 "loss": 0.0367,607 "reward": 0.2916666753590107,608 "reward_std": 0.24859581887722015,609 "rewards/accuracy_reward": 0.2916666753590107,610 "rewards/format_reward": 0.0,611 "step": 43612 },613 {614 "clip_ratio": 0.0,615 "completion_length": 424.91667556762695,616 "epoch": 0.4693333333333333,617 "grad_norm": 0.17440026998519897,618 "kl": 0.025390625,619 "learning_rate": 2.970022071578666e-06,620 "loss": 0.0344,621 "reward": 0.5833333432674408,622 "reward_std": 0.38552646338939667,623 "rewards/accuracy_reward": 0.5833333432674408,624 "rewards/format_reward": 0.0,625 "step": 44626 },627 {628 "clip_ratio": 0.0,629 "completion_length": 595.6041793823242,630 "epoch": 0.48,631 "grad_norm": 0.691380500793457,632 "kl": 0.03966522216796875,633 "learning_rate": 2.9661723160448267e-06,634 "loss": 0.0369,635 "reward": 0.18750000186264515,636 "reward_std": 0.11558076366782188,637 "rewards/accuracy_reward": 0.18750000186264515,638 "rewards/format_reward": 0.0,639 "step": 45640 },641 {642 "clip_ratio": 0.0,643 "completion_length": 583.2291870117188,644 "epoch": 0.49066666666666664,645 "grad_norm": 0.10043969005346298,646 "kl": 0.016597747802734375,647 "learning_rate": 2.962092876032427e-06,648 "loss": 0.0466,649 "reward": 0.3958333358168602,650 "reward_std": 0.2350771240890026,651 "rewards/accuracy_reward": 0.3958333358168602,652 "rewards/format_reward": 0.0,653 "step": 46654 },655 {656 "clip_ratio": 0.0,657 "completion_length": 535.3750152587891,658 "epoch": 0.5013333333333333,659 "grad_norm": 0.2007218450307846,660 "kl": 0.025909423828125,661 "learning_rate": 2.957784390609632e-06,662 "loss": -0.069,663 "reward": 0.4375000111758709,664 "reward_std": 0.3816108666360378,665 "rewards/accuracy_reward": 0.4375000111758709,666 "rewards/format_reward": 0.0,667 "step": 47668 },669 {670 "clip_ratio": 0.0,671 "completion_length": 489.0833435058594,672 "epoch": 0.512,673 "grad_norm": 0.1545257568359375,674 "kl": 0.03205108642578125,675 "learning_rate": 2.9532475347259113e-06,676 "loss": 0.0074,677 "reward": 0.4791666753590107,678 "reward_std": 0.3170611374080181,679 "rewards/accuracy_reward": 0.4791666753590107,680 "rewards/format_reward": 0.0,681 "step": 48682 },683 {684 "clip_ratio": 0.0,685 "completion_length": 518.416675567627,686 "epoch": 0.5226666666666666,687 "grad_norm": 0.15209315717220306,688 "kl": 0.0297698974609375,689 "learning_rate": 2.9484830191063057e-06,690 "loss": 0.0626,691 "reward": 0.47916667722165585,692 "reward_std": 0.299626849591732,693 "rewards/accuracy_reward": 0.47916667722165585,694 "rewards/format_reward": 0.0,695 "step": 49696 },697 {698 "clip_ratio": 0.0,699 "completion_length": 587.9375152587891,700 "epoch": 0.5333333333333333,701 "grad_norm": 0.14053413271903992,702 "kl": 0.023651123046875,703 "learning_rate": 2.9434915901400857e-06,704 "loss": 0.0446,705 "reward": 0.3125000074505806,706 "reward_std": 0.36417657881975174,707 "rewards/accuracy_reward": 0.3125000074505806,708 "rewards/format_reward": 0.0,709 "step": 50710 },711 {712 "clip_ratio": 0.0,713 "completion_length": 533.2083415985107,714 "epoch": 0.544,715 "grad_norm": 0.14134930074214935,716 "kl": 0.024688720703125,717 "learning_rate": 2.938274029763826e-06,718 "loss": 0.0683,719 "reward": 0.4166666753590107,720 "reward_std": 0.2861081659793854,721 "rewards/accuracy_reward": 0.4166666753590107,722 "rewards/format_reward": 0.0,723 "step": 51724 },725 {726 "clip_ratio": 0.0,727 "completion_length": 601.0625152587891,728 "epoch": 0.5546666666666666,729 "grad_norm": 0.14649143815040588,730 "kl": 0.02349853515625,731 "learning_rate": 2.9328311553389116e-06,732 "loss": 0.0642,733 "reward": 0.22916667349636555,734 "reward_std": 0.3092299550771713,735 "rewards/accuracy_reward": 0.22916667349636555,736 "rewards/format_reward": 0.0,737 "step": 52738 },739 {740 "clip_ratio": 0.0,741 "completion_length": 562.3541793823242,742 "epoch": 0.5653333333333334,743 "grad_norm": 0.1704944372177124,744 "kl": 0.0292816162109375,745 "learning_rate": 2.927163819523492e-06,746 "loss": -0.0439,747 "reward": 0.2916666753590107,748 "reward_std": 0.2861081585288048,749 "rewards/accuracy_reward": 0.2916666753590107,750 "rewards/format_reward": 0.0,751 "step": 53752 },753 {754 "clip_ratio": 0.0,755 "completion_length": 487.8541831970215,756 "epoch": 0.576,757 "grad_norm": 0.20558209717273712,758 "kl": 0.04265594482421875,759 "learning_rate": 2.9212729101389084e-06,760 "loss": 0.0393,761 "reward": 0.4166666828095913,762 "reward_std": 0.3332235999405384,763 "rewards/accuracy_reward": 0.4166666828095913,764 "rewards/format_reward": 0.0,765 "step": 54766 },767 {768 "clip_ratio": 0.0,769 "completion_length": 638.2500228881836,770 "epoch": 0.5866666666666667,771 "grad_norm": 0.2073916643857956,772 "kl": 0.02713775634765625,773 "learning_rate": 2.9151593500306103e-06,774 "loss": 0.043,775 "reward": 0.2708333395421505,776 "reward_std": 0.2621145099401474,777 "rewards/accuracy_reward": 0.2708333395421505,778 "rewards/format_reward": 0.0,779 "step": 55780 },781 {782 "clip_ratio": 0.0,783 "completion_length": 432.4791717529297,784 "epoch": 0.5973333333333334,785 "grad_norm": 0.14717544615268707,786 "kl": 0.030670166015625,787 "learning_rate": 2.9088240969235863e-06,788 "loss": 0.012,789 "reward": 0.4791666753590107,790 "reward_std": 0.42872630804777145,791 "rewards/accuracy_reward": 0.4791666753590107,792 "rewards/format_reward": 0.0,793 "step": 56794 },795 {796 "clip_ratio": 0.0,797 "completion_length": 434.2291793823242,798 "epoch": 0.608,799 "grad_norm": 0.1894180178642273,800 "kl": 0.069061279296875,801 "learning_rate": 2.902268143272333e-06,802 "loss": 0.0308,803 "reward": 0.6666666697710752,804 "reward_std": 0.24859581515192986,805 "rewards/accuracy_reward": 0.6666666697710752,806 "rewards/format_reward": 0.0,807 "step": 57808 },809 {810 "clip_ratio": 0.0,811 "completion_length": 467.66667556762695,812 "epoch": 0.6186666666666667,813 "grad_norm": 0.11211913079023361,814 "kl": 0.0320892333984375,815 "learning_rate": 2.895492516105378e-06,816 "loss": 0.0359,817 "reward": 0.4583333395421505,818 "reward_std": 0.22155842557549477,819 "rewards/accuracy_reward": 0.4583333395421505,820 "rewards/format_reward": 0.0,821 "step": 58822 },823 {824 "clip_ratio": 0.0,825 "completion_length": 611.6875076293945,826 "epoch": 0.6293333333333333,827 "grad_norm": 0.12812590599060059,828 "kl": 0.03234100341796875,829 "learning_rate": 2.888498276864393e-06,830 "loss": 0.0159,831 "reward": 0.3125000074505806,832 "reward_std": 0.2621145099401474,833 "rewards/accuracy_reward": 0.3125000074505806,834 "rewards/format_reward": 0.0,835 "step": 59836 },837 {838 "clip_ratio": 0.0,839 "completion_length": 699.4583473205566,840 "epoch": 0.64,841 "grad_norm": 0.0962519571185112,842 "kl": 0.02121734619140625,843 "learning_rate": 2.8812865212379087e-06,844 "loss": -0.016,845 "reward": 0.22916666977107525,846 "reward_std": 0.1801304928958416,847 "rewards/accuracy_reward": 0.22916666977107525,848 "rewards/format_reward": 0.0,849 "step": 60850 },851 {852 "clip_ratio": 0.0,853 "completion_length": 537.1458435058594,854 "epoch": 0.6506666666666666,855 "grad_norm": 0.4079749286174774,856 "kl": 0.03366851806640625,857 "learning_rate": 2.8738583789896743e-06,858 "loss": -0.0003,859 "reward": 0.3333333395421505,860 "reward_std": 0.3680921792984009,861 "rewards/accuracy_reward": 0.3333333395421505,862 "rewards/format_reward": 0.0,863 "step": 61864 },865 {866 "clip_ratio": 0.0,867 "completion_length": 590.4583435058594,868 "epoch": 0.6613333333333333,869 "grad_norm": 0.1474791318178177,870 "kl": 0.0286712646484375,871 "learning_rate": 2.866215013781668e-06,872 "loss": 0.0456,873 "reward": 0.20833334140479565,874 "reward_std": 0.268673874437809,875 "rewards/accuracy_reward": 0.20833334140479565,876 "rewards/format_reward": 0.0,877 "step": 62878 },879 {880 "clip_ratio": 0.0,881 "completion_length": 505.1250190734863,882 "epoch": 0.672,883 "grad_norm": 0.20406241714954376,884 "kl": 0.0316925048828125,885 "learning_rate": 2.858357622991806e-06,886 "loss": 0.0455,887 "reward": 0.4375000111758709,888 "reward_std": 0.41912321001291275,889 "rewards/accuracy_reward": 0.4375000111758709,890 "rewards/format_reward": 0.0,891 "step": 63892 },893 {894 "clip_ratio": 0.0,895 "completion_length": 592.6250228881836,896 "epoch": 0.6826666666666666,897 "grad_norm": 0.17144007980823517,898 "kl": 0.04378509521484375,899 "learning_rate": 2.8502874375263646e-06,900 "loss": 0.0515,901 "reward": 0.25000000186264515,902 "reward_std": 0.22155842557549477,903 "rewards/accuracy_reward": 0.25000000186264515,904 "rewards/format_reward": 0.0,905 "step": 64906 },907 {908 "clip_ratio": 0.0,909 "completion_length": 491.75001430511475,910 "epoch": 0.6933333333333334,911 "grad_norm": 0.21236185729503632,912 "kl": 0.0411224365234375,913 "learning_rate": 2.8420057216271533e-06,914 "loss": 0.07,915 "reward": 0.4166666716337204,916 "reward_std": 0.26603010296821594,917 "rewards/accuracy_reward": 0.4166666716337204,918 "rewards/format_reward": 0.0,919 "step": 65920 },921 {922 "clip_ratio": 0.0,923 "completion_length": 524.0000114440918,924 "epoch": 0.704,925 "grad_norm": 0.15796327590942383,926 "kl": 0.03389739990234375,927 "learning_rate": 2.8335137726734607e-06,928 "loss": 0.0943,929 "reward": 0.39583334140479565,930 "reward_std": 0.3720077648758888,931 "rewards/accuracy_reward": 0.39583334140479565,932 "rewards/format_reward": 0.0,933 "step": 66934 },935 {936 "clip_ratio": 0.0,937 "completion_length": 500.3125114440918,938 "epoch": 0.7146666666666667,939 "grad_norm": 0.27333250641822815,940 "kl": 0.04351806640625,941 "learning_rate": 2.8248129209788164e-06,942 "loss": 0.0156,943 "reward": 0.4375000149011612,944 "reward_std": 0.36417657881975174,945 "rewards/accuracy_reward": 0.4375000149011612,946 "rewards/format_reward": 0.0,947 "step": 67948 },949 {950 "clip_ratio": 0.0,951 "completion_length": 581.8750228881836,952 "epoch": 0.7253333333333334,953 "grad_norm": 4.9572272300720215,954 "kl": 0.10577392578125,955 "learning_rate": 2.8159045295825857e-06,956 "loss": 0.075,957 "reward": 0.2291666679084301,958 "reward_std": 0.2699456959962845,959 "rewards/accuracy_reward": 0.2291666679084301,960 "rewards/format_reward": 0.0,961 "step": 68962 },963 {964 "clip_ratio": 0.0,965 "completion_length": 510.7916793823242,966 "epoch": 0.736,967 "grad_norm": 0.2626616060733795,968 "kl": 0.047515869140625,969 "learning_rate": 2.806789994036445e-06,970 "loss": -0.0249,971 "reward": 0.479166679084301,972 "reward_std": 0.48367294296622276,973 "rewards/accuracy_reward": 0.479166679084301,974 "rewards/format_reward": 0.0,975 "step": 69976 },977 {978 "clip_ratio": 0.0,979 "completion_length": 478.4166793823242,980 "epoch": 0.7466666666666667,981 "grad_norm": 0.12211908400058746,982 "kl": 0.03977203369140625,983 "learning_rate": 2.7974707421857584e-06,984 "loss": 0.0285,985 "reward": 0.4166666679084301,986 "reward_std": 0.23899272084236145,987 "rewards/accuracy_reward": 0.4166666679084301,988 "rewards/format_reward": 0.0,989 "step": 70990 },991 {992 "clip_ratio": 0.0,993 "completion_length": 528.9583492279053,994 "epoch": 0.7573333333333333,995 "grad_norm": 0.1832711100578308,996 "kl": 0.056304931640625,997 "learning_rate": 2.7879482339458973e-06,998 "loss": -0.011,999 "reward": 0.3333333432674408,1000 "reward_std": 0.31314554437994957,1001 "rewards/accuracy_reward": 0.3333333432674408,1002 "rewards/format_reward": 0.0,1003 "step": 711004 },1005 {1006 "clip_ratio": 0.0,1007 "completion_length": 501.89584732055664,1008 "epoch": 0.768,1009 "grad_norm": 0.19136710464954376,1010 "kl": 0.0486297607421875,1011 "learning_rate": 2.778223961073537e-06,1012 "loss": 0.1018,1013 "reward": 0.3541666753590107,1014 "reward_std": 0.3170611411333084,1015 "rewards/accuracy_reward": 0.3541666753590107,1016 "rewards/format_reward": 0.0,1017 "step": 721018 },1019 {1020 "clip_ratio": 0.0,1021 "completion_length": 417.6458435058594,1022 "epoch": 0.7786666666666666,1023 "grad_norm": 0.17803442478179932,1024 "kl": 0.04766845703125,1025 "learning_rate": 2.7682994469329646e-06,1026 "loss": 0.0162,1027 "reward": 0.4375000074505806,1028 "reward_std": 0.36417658254504204,1029 "rewards/accuracy_reward": 0.4375000074505806,1030 "rewards/format_reward": 0.0,1031 "step": 731032 },1033 {1034 "clip_ratio": 0.0,1035 "completion_length": 545.5625190734863,1036 "epoch": 0.7893333333333333,1037 "grad_norm": 0.2556670606136322,1038 "kl": 0.0534820556640625,1039 "learning_rate": 2.7581762462574334e-06,1040 "loss": 0.0717,1041 "reward": 0.43750000931322575,1042 "reward_std": 0.2996268533170223,1043 "rewards/accuracy_reward": 0.43750000931322575,1044 "rewards/format_reward": 0.0,1045 "step": 741046 },1047 {1048 "clip_ratio": 0.0,1049 "completion_length": 727.8125152587891,1050 "epoch": 0.8,1051 "grad_norm": 2.5011374950408936,1052 "kl": 0.15460205078125,1053 "learning_rate": 2.747855944905606e-06,1054 "loss": 0.0504,1055 "reward": 0.25000000558793545,1056 "reward_std": 0.3506578877568245,1057 "rewards/accuracy_reward": 0.25000000558793545,1058 "rewards/format_reward": 0.0,1059 "step": 751060 },1061 {1062 "clip_ratio": 0.0,1063 "completion_length": 559.6875152587891,1064 "epoch": 0.8106666666666666,1065 "grad_norm": 0.18483959138393402,1066 "kl": 0.0756072998046875,1067 "learning_rate": 2.73734015961312e-06,1068 "loss": 0.0674,1069 "reward": 0.33333334140479565,1070 "reward_std": 0.3776952773332596,1071 "rewards/accuracy_reward": 0.33333334140479565,1072 "rewards/format_reward": 0.0,1073 "step": 761074 },1075 {1076 "clip_ratio": 0.0,1077 "completion_length": 569.7708511352539,1078 "epoch": 0.8213333333333334,1079 "grad_norm": 0.23977096378803253,1080 "kl": 0.077392578125,1081 "learning_rate": 2.7266305377393183e-06,1082 "loss": -0.0037,1083 "reward": 0.37500001303851604,1084 "reward_std": 0.3977733291685581,1085 "rewards/accuracy_reward": 0.37500001303851604,1086 "rewards/format_reward": 0.0,1087 "step": 771088 },1089 {1090 "clip_ratio": 0.0,1091 "completion_length": 673.3541870117188,1092 "epoch": 0.832,1093 "grad_norm": 0.15160395205020905,1094 "kl": 0.087005615234375,1095 "learning_rate": 2.7157287570091765e-06,1096 "loss": 0.0068,1097 "reward": 0.20833333767950535,1098 "reward_std": 0.3506578840315342,1099 "rewards/accuracy_reward": 0.20833333767950535,1100 "rewards/format_reward": 0.0,1101 "step": 781102 },1103 {1104 "clip_ratio": 0.0,1105 "completion_length": 665.2916870117188,1106 "epoch": 0.8426666666666667,1107 "grad_norm": 0.18723702430725098,1108 "kl": 0.086883544921875,1109 "learning_rate": 2.7046365252504835e-06,1110 "loss": 0.1126,1111 "reward": 0.2916666716337204,1112 "reward_std": 0.26603010296821594,1113 "rewards/accuracy_reward": 0.2916666716337204,1114 "rewards/format_reward": 0.0,1115 "step": 791116 },1117 {1118 "clip_ratio": 0.0,1119 "completion_length": 570.3750228881836,1120 "epoch": 0.8533333333333334,1121 "grad_norm": 0.20665200054645538,1122 "kl": 0.130096435546875,1123 "learning_rate": 2.693355580126295e-06,1124 "loss": 0.043,1125 "reward": 0.31250000931322575,1126 "reward_std": 0.2996268533170223,1127 "rewards/accuracy_reward": 0.31250000931322575,1128 "rewards/format_reward": 0.0,1129 "step": 801130 },1131 {1132 "clip_ratio": 0.0,1133 "completion_length": 474.5625114440918,1134 "epoch": 0.864,1135 "grad_norm": 0.17555448412895203,1136 "kl": 0.089813232421875,1137 "learning_rate": 2.6818876888627203e-06,1138 "loss": -0.0371,1139 "reward": 0.3750000037252903,1140 "reward_std": 0.16661180183291435,1141 "rewards/accuracy_reward": 0.3750000037252903,1142 "rewards/format_reward": 0.0,1143 "step": 811144 },1145 {1146 "clip_ratio": 0.0,1147 "completion_length": 517.9583530426025,1148 "epoch": 0.8746666666666667,1149 "grad_norm": 0.24799272418022156,1150 "kl": 0.057342529296875,1151 "learning_rate": 2.6702346479720777e-06,1152 "loss": 0.1192,1153 "reward": 0.41666667722165585,1154 "reward_std": 0.22155843675136566,1155 "rewards/accuracy_reward": 0.41666667722165585,1156 "rewards/format_reward": 0.0,1157 "step": 821158 },1159 {1160 "clip_ratio": 0.0,1161 "completion_length": 539.0000114440918,1162 "epoch": 0.8853333333333333,1163 "grad_norm": 0.18812210857868195,1164 "kl": 0.085723876953125,1165 "learning_rate": 2.6583982829714576e-06,1166 "loss": 0.1357,1167 "reward": 0.47916667722165585,1168 "reward_std": 0.36417657881975174,1169 "rewards/accuracy_reward": 0.47916667722165585,1170 "rewards/format_reward": 0.0,1171 "step": 831172 },1173 {1174 "clip_ratio": 0.0,1175 "completion_length": 599.6875114440918,1176 "epoch": 0.896,1177 "grad_norm": 0.15599775314331055,1178 "kl": 0.13250732421875,1179 "learning_rate": 2.6463804480967447e-06,1180 "loss": 0.1373,1181 "reward": 0.20833334140479565,1182 "reward_std": 0.268673874437809,1183 "rewards/accuracy_reward": 0.20833334140479565,1184 "rewards/format_reward": 0.0,1185 "step": 841186 },1187 {1188 "clip_ratio": 0.0,1189 "completion_length": 623.2916831970215,1190 "epoch": 0.9066666666666666,1191 "grad_norm": 0.1542806178331375,1192 "kl": 0.14483642578125,1193 "learning_rate": 2.6341830260121445e-06,1194 "loss": 0.0449,1195 "reward": 0.2291666679084301,1196 "reward_std": 0.1705273911356926,1197 "rewards/accuracy_reward": 0.2291666679084301,1198 "rewards/format_reward": 0.0,1199 "step": 851200 },