GD-ML/Video-STAR-7B
169
1{2 "best_metric": null,3 "best_model_checkpoint": null,4 "epoch": 1.0,5 "eval_steps": 500,6 "global_step": 585,7 "is_hyper_param_search": false,8 "is_local_process_zero": true,9 "is_world_process_zero": true,10 "log_history": [11 {12 "all_correct": 0.625,13 "all_wrong": 0.0,14 "completion_length": 360.09375,15 "epoch": 0.0017094017094017094,16 "grad_norm": 1.9826301268041722,17 "kl": 0.0,18 "learning_rate": 4.999963950687845e-07,19 "loss": 0.0,20 "reward": 2.6583333015441895,21 "reward_std": 0.8285301923751831,22 "rewards/accuracy_reward": 1.7989583015441895,23 "rewards/format_reward": 0.53125,24 "step": 1,25 "temporal_rewards": 0.7526 },27 {28 "all_correct": 0.75,29 "all_wrong": 0.0,30 "completion_length": 460.71875,31 "epoch": 0.003418803418803419,32 "grad_norm": 1.783036944219682,33 "kl": 0.00041484832763671875,34 "learning_rate": 4.999855803791026e-07,35 "loss": 0.0,36 "reward": 3.078125,37 "reward_std": 0.714574933052063,38 "rewards/accuracy_reward": 2.090625047683716,39 "rewards/format_reward": 0.65625,40 "step": 2,41 "temporal_rewards": 0.7542 },43 {44 "all_correct": 0.75,45 "all_wrong": 0.0,46 "completion_length": 458.53125,47 "epoch": 0.005128205128205128,48 "grad_norm": 1.9140893572712068,49 "kl": 0.004611968994140625,50 "learning_rate": 4.999675562428436e-07,51 "loss": 0.0002,52 "reward": 3.1812500953674316,53 "reward_std": 0.6964144110679626,54 "rewards/accuracy_reward": 2.0531249046325684,55 "rewards/format_reward": 0.78125,56 "step": 3,57 "temporal_rewards": 0.87558 },59 {60 "all_correct": 0.625,61 "all_wrong": 0.125,62 "completion_length": 342.34375,63 "epoch": 0.006837606837606838,64 "grad_norm": 1.6435177595568693,65 "kl": 0.0047607421875,66 "learning_rate": 4.99942323179814e-07,67 "loss": 0.0002,68 "reward": 3.0593748092651367,69 "reward_std": 0.4561102092266083,70 "rewards/accuracy_reward": 1.834375023841858,71 "rewards/format_reward": 0.9375,72 "step": 4,73 "temporal_rewards": 0.62574 },75 {76 "all_correct": 0.75,77 "all_wrong": 0.0,78 "completion_length": 369.84375,79 "epoch": 0.008547008547008548,80 "grad_norm": 1.9133739405680183,81 "kl": 0.004207611083984375,82 "learning_rate": 4.999098819177214e-07,83 "loss": 0.0002,84 "reward": 3.379166603088379,85 "reward_std": 0.6761210560798645,86 "rewards/accuracy_reward": 2.0791666507720947,87 "rewards/format_reward": 0.9375,88 "step": 5,89 "temporal_rewards": 0.7590 },91 {92 "all_correct": 0.625,93 "all_wrong": 0.0,94 "completion_length": 351.4375,95 "epoch": 0.010256410256410256,96 "grad_norm": 1.7373006031428762,97 "kl": 0.008636474609375,98 "learning_rate": 4.998702333921537e-07,99 "loss": 0.0003,100 "reward": 3.2947916984558105,101 "reward_std": 0.7285541296005249,102 "rewards/accuracy_reward": 1.988541603088379,103 "rewards/format_reward": 1.0,104 "step": 6,105 "temporal_rewards": 0.75106 },107 {108 "all_correct": 0.875,109 "all_wrong": 0.0,110 "completion_length": 330.90625,111 "epoch": 0.011965811965811967,112 "grad_norm": 1.81484083295842,113 "kl": 0.00798797607421875,114 "learning_rate": 4.998233787465529e-07,115 "loss": 0.0003,116 "reward": 3.910416603088379,117 "reward_std": 0.41429048776626587,118 "rewards/accuracy_reward": 2.519791603088379,119 "rewards/format_reward": 1.0,120 "step": 7,121 "temporal_rewards": 1.0122 },123 {124 "all_correct": 0.75,125 "all_wrong": 0.125,126 "completion_length": 337.34375,127 "epoch": 0.013675213675213675,128 "grad_norm": 1.5267078479800362,129 "kl": 0.00785064697265625,130 "learning_rate": 4.99769319332181e-07,131 "loss": 0.0003,132 "reward": 3.3531250953674316,133 "reward_std": 0.3145015239715576,134 "rewards/accuracy_reward": 2.034374952316284,135 "rewards/format_reward": 1.0,136 "step": 8,137 "temporal_rewards": 0.75138 },139 {140 "all_correct": 0.875,141 "all_wrong": 0.0,142 "completion_length": 335.0625,143 "epoch": 0.015384615384615385,144 "grad_norm": 1.685582407194424,145 "kl": 0.00934600830078125,146 "learning_rate": 4.997080567080816e-07,147 "loss": 0.0004,148 "reward": 3.6895830631256104,149 "reward_std": 0.36123037338256836,150 "rewards/accuracy_reward": 2.3427083492279053,151 "rewards/format_reward": 0.96875,152 "step": 9,153 "temporal_rewards": 1.0154 },155 {156 "all_correct": 0.875,157 "all_wrong": 0.0,158 "completion_length": 353.875,159 "epoch": 0.017094017094017096,160 "grad_norm": 1.5329400178924226,161 "kl": 0.0093841552734375,162 "learning_rate": 4.996395926410354e-07,163 "loss": 0.0004,164 "reward": 3.886458396911621,165 "reward_std": 0.3707513213157654,166 "rewards/accuracy_reward": 2.4739582538604736,167 "rewards/format_reward": 0.96875,168 "step": 10,169 "temporal_rewards": 1.0170 },171 {172 "all_correct": 1.0,173 "all_wrong": 0.0,174 "completion_length": 351.0,175 "epoch": 0.018803418803418803,176 "grad_norm": 1.549766282148727,177 "kl": 0.0126953125,178 "learning_rate": 4.995639291055083e-07,179 "loss": 0.0005,180 "reward": 4.042708396911621,181 "reward_std": 0.2290562391281128,182 "rewards/accuracy_reward": 2.586458206176758,183 "rewards/format_reward": 1.0,184 "step": 11,185 "temporal_rewards": 1.0186 },187 {188 "all_correct": 0.75,189 "all_wrong": 0.0,190 "completion_length": 358.1875,191 "epoch": 0.020512820512820513,192 "grad_norm": 1.496305479107766,193 "kl": 0.016632080078125,194 "learning_rate": 4.994810682835951e-07,195 "loss": 0.0007,196 "reward": 3.6510417461395264,197 "reward_std": 0.5492165088653564,198 "rewards/accuracy_reward": 2.3135416507720947,199 "rewards/format_reward": 0.96875,200 "step": 12,201 "temporal_rewards": 0.75202 },203 {204 "all_correct": 0.75,205 "all_wrong": 0.0,206 "completion_length": 368.4375,207 "epoch": 0.022222222222222223,208 "grad_norm": 1.541763401661613,209 "kl": 0.0149078369140625,210 "learning_rate": 4.99391012564956e-07,211 "loss": 0.0006,212 "reward": 3.7291665077209473,213 "reward_std": 0.4612278342247009,214 "rewards/accuracy_reward": 2.351041793823242,215 "rewards/format_reward": 0.96875,216 "step": 13,217 "temporal_rewards": 0.875218 },219 {220 "all_correct": 0.875,221 "all_wrong": 0.0,222 "completion_length": 374.09375,223 "epoch": 0.023931623931623933,224 "grad_norm": 1.548862572081062,225 "kl": 0.0169219970703125,226 "learning_rate": 4.992937645467487e-07,227 "loss": 0.0007,228 "reward": 3.936458110809326,229 "reward_std": 0.32494786381721497,230 "rewards/accuracy_reward": 2.5614583492279053,231 "rewards/format_reward": 0.96875,232 "step": 14,233 "temporal_rewards": 0.875234 },235 {236 "all_correct": 0.875,237 "all_wrong": 0.0,238 "completion_length": 374.21875,239 "epoch": 0.02564102564102564,240 "grad_norm": 1.3211685700244122,241 "kl": 0.0189056396484375,242 "learning_rate": 4.991893270335525e-07,243 "loss": 0.0008,244 "reward": 3.8302083015441895,245 "reward_std": 0.275448203086853,246 "rewards/accuracy_reward": 2.386458396911621,247 "rewards/format_reward": 1.0,248 "step": 15,249 "temporal_rewards": 1.0250 },251 {252 "all_correct": 0.875,253 "all_wrong": 0.0,254 "completion_length": 390.65625,255 "epoch": 0.02735042735042735,256 "grad_norm": 1.0827777389563216,257 "kl": 0.0201568603515625,258 "learning_rate": 4.990777030372877e-07,259 "loss": 0.0008,260 "reward": 3.918750047683716,261 "reward_std": 0.27283355593681335,262 "rewards/accuracy_reward": 2.46875,263 "rewards/format_reward": 1.0,264 "step": 16,265 "temporal_rewards": 1.0266 },267 {268 "all_correct": 0.875,269 "all_wrong": 0.0,270 "completion_length": 396.5,271 "epoch": 0.02905982905982906,272 "grad_norm": 1.4719325483552113,273 "kl": 0.0209808349609375,274 "learning_rate": 4.989588957771289e-07,275 "loss": 0.0008,276 "reward": 4.057291030883789,277 "reward_std": 0.38743653893470764,278 "rewards/accuracy_reward": 2.613541603088379,279 "rewards/format_reward": 1.0,280 "step": 17,281 "temporal_rewards": 0.875282 },283 {284 "all_correct": 1.0,285 "all_wrong": 0.0,286 "completion_length": 377.15625,287 "epoch": 0.03076923076923077,288 "grad_norm": 1.3897844926570597,289 "kl": 0.0229339599609375,290 "learning_rate": 4.988329086794122e-07,291 "loss": 0.0009,292 "reward": 4.282291412353516,293 "reward_std": 0.2059958279132843,294 "rewards/accuracy_reward": 2.8260416984558105,295 "rewards/format_reward": 1.0,296 "step": 18,297 "temporal_rewards": 1.0298 },299 {300 "all_correct": 0.875,301 "all_wrong": 0.0,302 "completion_length": 394.75,303 "epoch": 0.03247863247863248,304 "grad_norm": 1.4025810108116477,305 "kl": 0.0246124267578125,306 "learning_rate": 4.98699745377536e-07,307 "loss": 0.001,308 "reward": 3.6979165077209473,309 "reward_std": 0.22571580111980438,310 "rewards/accuracy_reward": 2.285416603088379,311 "rewards/format_reward": 1.0,312 "step": 19,313 "temporal_rewards": 0.875314 },315 {316 "all_correct": 0.875,317 "all_wrong": 0.0,318 "completion_length": 395.375,319 "epoch": 0.03418803418803419,320 "grad_norm": 1.3422743689560948,321 "kl": 0.0244293212890625,322 "learning_rate": 4.98559409711857e-07,323 "loss": 0.001,324 "reward": 4.1895833015441895,325 "reward_std": 0.34896954894065857,326 "rewards/accuracy_reward": 2.764583110809326,327 "rewards/format_reward": 1.0,328 "step": 20,329 "temporal_rewards": 0.875330 },331 {332 "all_correct": 1.0,333 "all_wrong": 0.0,334 "completion_length": 391.5625,335 "epoch": 0.035897435897435895,336 "grad_norm": 1.3706326401655777,337 "kl": 0.0267791748046875,338 "learning_rate": 4.984119057295782e-07,339 "loss": 0.0011,340 "reward": 4.2083330154418945,341 "reward_std": 0.21551749110221863,342 "rewards/accuracy_reward": 2.714583396911621,343 "rewards/format_reward": 1.0,344 "step": 21,345 "temporal_rewards": 1.0346 },347 {348 "all_correct": 0.875,349 "all_wrong": 0.0,350 "completion_length": 416.375,351 "epoch": 0.037606837606837605,352 "grad_norm": 1.3430039575658776,353 "kl": 0.0264739990234375,354 "learning_rate": 4.982572376846336e-07,355 "loss": 0.0011,356 "reward": 4.012499809265137,357 "reward_std": 0.36676347255706787,358 "rewards/accuracy_reward": 2.5687501430511475,359 "rewards/format_reward": 1.0,360 "step": 22,361 "temporal_rewards": 0.875362 },363 {364 "all_correct": 1.0,365 "all_wrong": 0.0,366 "completion_length": 414.5,367 "epoch": 0.039316239316239315,368 "grad_norm": 1.0704324370355776,369 "kl": 0.027374267578125,370 "learning_rate": 4.980954100375641e-07,371 "loss": 0.0011,372 "reward": 3.9510416984558105,373 "reward_std": 0.21451421082019806,374 "rewards/accuracy_reward": 2.476041793823242,375 "rewards/format_reward": 1.0,376 "step": 23,377 "temporal_rewards": 1.0378 },379 {380 "all_correct": 0.75,381 "all_wrong": 0.125,382 "completion_length": 385.6875,383 "epoch": 0.041025641025641026,384 "grad_norm": 1.319994732732891,385 "kl": 0.0298309326171875,386 "learning_rate": 4.979264274553905e-07,387 "loss": 0.0012,388 "reward": 3.832291603088379,389 "reward_std": 0.3102354407310486,390 "rewards/accuracy_reward": 2.4510416984558105,391 "rewards/format_reward": 1.0,392 "step": 24,393 "temporal_rewards": 0.75394 },395 {396 "all_correct": 1.0,397 "all_wrong": 0.0,398 "completion_length": 417.15625,399 "epoch": 0.042735042735042736,400 "grad_norm": 1.2440324072815687,401 "kl": 0.025665283203125,402 "learning_rate": 4.977502948114771e-07,403 "loss": 0.001,404 "reward": 4.2677083015441895,405 "reward_std": 0.11905767768621445,406 "rewards/accuracy_reward": 2.7989585399627686,407 "rewards/format_reward": 1.0,408 "step": 25,409 "temporal_rewards": 1.0410 },411 {412 "all_correct": 0.875,413 "all_wrong": 0.0,414 "completion_length": 397.78125,415 "epoch": 0.044444444444444446,416 "grad_norm": 1.3286663602585649,417 "kl": 0.02685546875,418 "learning_rate": 4.975670171853925e-07,419 "loss": 0.0011,420 "reward": 3.9354166984558105,421 "reward_std": 0.29135236144065857,422 "rewards/accuracy_reward": 2.522916793823242,423 "rewards/format_reward": 1.0,424 "step": 26,425 "temporal_rewards": 0.875426 },427 {428 "all_correct": 0.875,429 "all_wrong": 0.0,430 "completion_length": 429.0,431 "epoch": 0.046153846153846156,432 "grad_norm": 1.301323110823831,433 "kl": 0.0247650146484375,434 "learning_rate": 4.973765998627628e-07,435 "loss": 0.001,436 "reward": 3.8374998569488525,437 "reward_std": 0.3257061839103699,438 "rewards/accuracy_reward": 2.4000000953674316,439 "rewards/format_reward": 1.0,440 "step": 27,441 "temporal_rewards": 0.875442 },443 {444 "all_correct": 1.0,445 "all_wrong": 0.0,446 "completion_length": 384.28125,447 "epoch": 0.04786324786324787,448 "grad_norm": 1.2554920983763502,449 "kl": 0.029266357421875,450 "learning_rate": 4.971790483351185e-07,451 "loss": 0.0012,452 "reward": 4.2864580154418945,453 "reward_std": 0.131798654794693,454 "rewards/accuracy_reward": 2.7989583015441895,455 "rewards/format_reward": 1.0,456 "step": 28,457 "temporal_rewards": 1.0458 },459 {460 "all_correct": 0.75,461 "all_wrong": 0.0,462 "completion_length": 394.0625,463 "epoch": 0.04957264957264957,464 "grad_norm": 0.9339974580407308,465 "kl": 0.0288543701171875,466 "learning_rate": 4.969743682997371e-07,467 "loss": 0.0012,468 "reward": 3.6687498092651367,469 "reward_std": 0.45044267177581787,470 "rewards/accuracy_reward": 2.25,471 "rewards/format_reward": 1.0,472 "step": 29,473 "temporal_rewards": 0.875474 },475 {476 "all_correct": 1.0,477 "all_wrong": 0.0,478 "completion_length": 392.4375,479 "epoch": 0.05128205128205128,480 "grad_norm": 1.2262654726579296,481 "kl": 0.0308685302734375,482 "learning_rate": 4.967625656594781e-07,483 "loss": 0.0012,484 "reward": 4.2395830154418945,485 "reward_std": 0.16444088518619537,486 "rewards/accuracy_reward": 2.7708332538604736,487 "rewards/format_reward": 1.0,488 "step": 30,489 "temporal_rewards": 1.0490 },491 {492 "all_correct": 1.0,493 "all_wrong": 0.0,494 "completion_length": 381.25,495 "epoch": 0.05299145299145299,496 "grad_norm": 1.2573879496841676,497 "kl": 0.0293731689453125,498 "learning_rate": 4.965436465226134e-07,499 "loss": 0.0012,500 "reward": 4.016666412353516,501 "reward_std": 0.14875400066375732,502 "rewards/accuracy_reward": 2.535416603088379,503 "rewards/format_reward": 1.0,504 "step": 31,505 "temporal_rewards": 1.0506 },507 {508 "all_correct": 0.875,509 "all_wrong": 0.0,510 "completion_length": 368.6875,511 "epoch": 0.0547008547008547,512 "grad_norm": 1.2564985420337802,513 "kl": 0.030120849609375,514 "learning_rate": 4.963176172026501e-07,515 "loss": 0.0012,516 "reward": 3.9968748092651367,517 "reward_std": 0.3774029016494751,518 "rewards/accuracy_reward": 2.559375047683716,519 "rewards/format_reward": 1.0,520 "step": 32,521 "temporal_rewards": 0.875522 },523 {524 "all_correct": 0.875,525 "all_wrong": 0.0,526 "completion_length": 424.0625,527 "epoch": 0.05641025641025641,528 "grad_norm": 0.9694026424422826,529 "kl": 0.033355712890625,530 "learning_rate": 4.960844842181495e-07,531 "loss": 0.0013,532 "reward": 4.047916412353516,533 "reward_std": 0.3180779814720154,534 "rewards/accuracy_reward": 2.616666555404663,535 "rewards/format_reward": 1.0,536 "step": 33,537 "temporal_rewards": 0.875538 },539 {540 "all_correct": 1.0,541 "all_wrong": 0.0,542 "completion_length": 372.03125,543 "epoch": 0.05811965811965812,544 "grad_norm": 1.1514296295303623,545 "kl": 0.03143310546875,546 "learning_rate": 4.958442542925385e-07,547 "loss": 0.0013,548 "reward": 3.9666666984558105,549 "reward_std": 0.14849570393562317,550 "rewards/accuracy_reward": 2.4791667461395264,551 "rewards/format_reward": 1.0,552 "step": 34,553 "temporal_rewards": 1.0554 },555 {556 "all_correct": 1.0,557 "all_wrong": 0.0,558 "completion_length": 362.71875,559 "epoch": 0.05982905982905983,560 "grad_norm": 1.1335397647489873,561 "kl": 0.0335693359375,562 "learning_rate": 4.955969343539162e-07,563 "loss": 0.0013,564 "reward": 4.120833396911621,565 "reward_std": 0.1538814902305603,566 "rewards/accuracy_reward": 2.6583333015441895,567 "rewards/format_reward": 1.0,568 "step": 35,569 "temporal_rewards": 1.0570 },571 {572 "all_correct": 0.875,573 "all_wrong": 0.0,574 "completion_length": 385.40625,575 "epoch": 0.06153846153846154,576 "grad_norm": 1.3288920319551032,577 "kl": 0.03155517578125,578 "learning_rate": 4.953425315348533e-07,579 "loss": 0.0013,580 "reward": 4.229166507720947,581 "reward_std": 0.2994784116744995,582 "rewards/accuracy_reward": 2.7760415077209473,583 "rewards/format_reward": 1.0,584 "step": 36,585 "temporal_rewards": 1.0586 },587 {588 "all_correct": 1.0,589 "all_wrong": 0.0,590 "completion_length": 407.46875,591 "epoch": 0.06324786324786325,592 "grad_norm": 1.3491529940946705,593 "kl": 0.0291900634765625,594 "learning_rate": 4.950810531721873e-07,595 "loss": 0.0012,596 "reward": 4.1583333015441895,597 "reward_std": 0.08171668648719788,598 "rewards/accuracy_reward": 2.727083206176758,599 "rewards/format_reward": 1.0,600 "step": 37,601 "temporal_rewards": 0.875602 },603 {604 "all_correct": 1.0,605 "all_wrong": 0.0,606 "completion_length": 378.375,607 "epoch": 0.06495726495726496,608 "grad_norm": 1.4603865780151248,609 "kl": 0.02972412109375,610 "learning_rate": 4.948125068068102e-07,611 "loss": 0.0012,612 "reward": 4.2864580154418945,613 "reward_std": 0.11952318251132965,614 "rewards/accuracy_reward": 2.811458110809326,615 "rewards/format_reward": 1.0,616 "step": 38,617 "temporal_rewards": 1.0618 },619 {620 "all_correct": 1.0,621 "all_wrong": 0.0,622 "completion_length": 476.25,623 "epoch": 0.06666666666666667,624 "grad_norm": 1.3332368328871798,625 "kl": 0.02838134765625,626 "learning_rate": 4.945369001834514e-07,627 "loss": 0.0011,628 "reward": 4.363541603088379,629 "reward_std": 0.16119801998138428,630 "rewards/accuracy_reward": 2.8885416984558105,631 "rewards/format_reward": 1.0,632 "step": 39,633 "temporal_rewards": 1.0634 },635 {636 "all_correct": 1.0,637 "all_wrong": 0.0,638 "completion_length": 382.90625,639 "epoch": 0.06837606837606838,640 "grad_norm": 1.1501096034585943,641 "kl": 0.0298004150390625,642 "learning_rate": 4.942542412504542e-07,643 "loss": 0.0012,644 "reward": 4.179166316986084,645 "reward_std": 0.09968569874763489,646 "rewards/accuracy_reward": 2.6854166984558105,647 "rewards/format_reward": 1.0,648 "step": 40,649 "temporal_rewards": 1.0650 },651 {652 "all_correct": 1.0,653 "all_wrong": 0.0,654 "completion_length": 377.78125,655 "epoch": 0.07008547008547009,656 "grad_norm": 1.3385150586201982,657 "kl": 0.0303802490234375,658 "learning_rate": 4.939645381595469e-07,659 "loss": 0.0012,660 "reward": 4.415625095367432,661 "reward_std": 0.12376689910888672,662 "rewards/accuracy_reward": 2.9468750953674316,663 "rewards/format_reward": 1.0,664 "step": 41,665 "temporal_rewards": 1.0666 },667 {668 "all_correct": 1.0,669 "all_wrong": 0.0,670 "completion_length": 370.125,671 "epoch": 0.07179487179487179,672 "grad_norm": 1.5092483520373827,673 "kl": 0.032257080078125,674 "learning_rate": 4.93667799265607e-07,675 "loss": 0.0013,676 "reward": 4.179166793823242,677 "reward_std": 0.24446402490139008,678 "rewards/accuracy_reward": 2.7291665077209473,679 "rewards/format_reward": 1.0,680 "step": 42,681 "temporal_rewards": 0.875682 },683 {684 "all_correct": 1.0,685 "all_wrong": 0.0,686 "completion_length": 413.6875,687 "epoch": 0.0735042735042735,688 "grad_norm": 0.8198637140300333,689 "kl": 0.0333251953125,690 "learning_rate": 4.933640331264209e-07,691 "loss": 0.0013,692 "reward": 4.356249809265137,693 "reward_std": 0.02693380042910576,694 "rewards/accuracy_reward": 2.862499952316284,695 "rewards/format_reward": 1.0,696 "step": 43,697 "temporal_rewards": 1.0698 },699 {700 "all_correct": 1.0,701 "all_wrong": 0.0,702 "completion_length": 377.34375,703 "epoch": 0.07521367521367521,704 "grad_norm": 1.362239639644523,705 "kl": 0.03790283203125,706 "learning_rate": 4.930532485024371e-07,707 "loss": 0.0015,708 "reward": 4.239583492279053,709 "reward_std": 0.15847893059253693,710 "rewards/accuracy_reward": 2.7645833492279053,711 "rewards/format_reward": 1.0,712 "step": 44,713 "temporal_rewards": 1.0714 },715 {716 "all_correct": 1.0,717 "all_wrong": 0.0,718 "completion_length": 383.34375,719 "epoch": 0.07692307692307693,720 "grad_norm": 1.348031471721414,721 "kl": 0.03143310546875,722 "learning_rate": 4.92735454356513e-07,723 "loss": 0.0013,724 "reward": 4.290624618530273,725 "reward_std": 0.0891985222697258,726 "rewards/accuracy_reward": 2.796875,727 "rewards/format_reward": 1.0,728 "step": 45,729 "temporal_rewards": 1.0730 },731 {732 "all_correct": 1.0,733 "all_wrong": 0.0,734 "completion_length": 393.8125,735 "epoch": 0.07863247863247863,736 "grad_norm": 1.0447875433549505,737 "kl": 0.0330810546875,738 "learning_rate": 4.924106598536569e-07,739 "loss": 0.0013,740 "reward": 4.339583396911621,741 "reward_std": 0.0612565279006958,742 "rewards/accuracy_reward": 2.858333110809326,743 "rewards/format_reward": 1.0,744 "step": 46,745 "temporal_rewards": 1.0746 },747 {748 "all_correct": 1.0,749 "all_wrong": 0.0,750 "completion_length": 394.03125,751 "epoch": 0.08034188034188035,752 "grad_norm": 1.1567990099669652,753 "kl": 0.03277587890625,754 "learning_rate": 4.920788743607635e-07,755 "loss": 0.0013,756 "reward": 4.3927083015441895,757 "reward_std": 0.13107708096504211,758 "rewards/accuracy_reward": 2.9114582538604736,759 "rewards/format_reward": 1.0,760 "step": 47,761 "temporal_rewards": 1.0762 },763 {764 "all_correct": 1.0,765 "all_wrong": 0.0,766 "completion_length": 403.8125,767 "epoch": 0.08205128205128205,768 "grad_norm": 1.0369718009838986,769 "kl": 0.03460693359375,770 "learning_rate": 4.917401074463441e-07,771 "loss": 0.0014,772 "reward": 4.4270830154418945,773 "reward_std": 0.13362948596477509,774 "rewards/accuracy_reward": 2.9270834922790527,775 "rewards/format_reward": 1.0,776 "step": 48,777 "temporal_rewards": 1.0778 },779 {780 "all_correct": 1.0,781 "all_wrong": 0.0,782 "completion_length": 392.25,783 "epoch": 0.08376068376068375,784 "grad_norm": 0.7093588876792561,785 "kl": 0.037353515625,786 "learning_rate": 4.913943688802497e-07,787 "loss": 0.0015,788 "reward": 3.8843750953674316,789 "reward_std": 0.1348297894001007,790 "rewards/accuracy_reward": 2.390625,791 "rewards/format_reward": 1.0,792 "step": 49,793 "temporal_rewards": 1.0794 },795 {796 "all_correct": 1.0,797 "all_wrong": 0.0,798 "completion_length": 415.625,799 "epoch": 0.08547008547008547,800 "grad_norm": 0.9797316320763304,801 "kl": 0.035736083984375,802 "learning_rate": 4.910416686333906e-07,803 "loss": 0.0014,804 "reward": 3.855208396911621,805 "reward_std": 0.03958335518836975,806 "rewards/accuracy_reward": 2.3614583015441895,807 "rewards/format_reward": 1.0,808 "step": 50,809 "temporal_rewards": 1.0810 },811 {812 "all_correct": 1.0,813 "all_wrong": 0.0,814 "completion_length": 402.71875,815 "epoch": 0.08717948717948718,816 "grad_norm": 0.7267487162340365,817 "kl": 0.034515380859375,818 "learning_rate": 4.906820168774477e-07,819 "loss": 0.0014,820 "reward": 4.229166507720947,821 "reward_std": 0.04166668653488159,822 "rewards/accuracy_reward": 2.7291665077209473,823 "rewards/format_reward": 1.0,824 "step": 51,825 "temporal_rewards": 1.0826 },827 {828 "all_correct": 1.0,829 "all_wrong": 0.0,830 "completion_length": 408.65625,831 "epoch": 0.08888888888888889,832 "grad_norm": 0.9894251393442087,833 "kl": 0.0360107421875,834 "learning_rate": 4.903154239845797e-07,835 "loss": 0.0014,836 "reward": 4.327083587646484,837 "reward_std": 0.09055028110742569,838 "rewards/accuracy_reward": 2.8270833492279053,839 "rewards/format_reward": 1.0,840 "step": 52,841 "temporal_rewards": 1.0842 },843 {844 "all_correct": 1.0,845 "all_wrong": 0.0,846 "completion_length": 403.21875,847 "epoch": 0.0905982905982906,848 "grad_norm": 1.2134526914387425,849 "kl": 0.0322265625,850 "learning_rate": 4.899419005271241e-07,851 "loss": 0.0013,852 "reward": 4.413541793823242,853 "reward_std": 0.1331278383731842,854 "rewards/accuracy_reward": 2.9197916984558105,855 "rewards/format_reward": 1.0,856 "step": 53,857 "temporal_rewards": 1.0858 },859 {860 "all_correct": 1.0,861 "all_wrong": 0.0,862 "completion_length": 428.34375,863 "epoch": 0.09230769230769231,864 "grad_norm": 1.1445171148845454,865 "kl": 0.035369873046875,866 "learning_rate": 4.895614572772916e-07,867 "loss": 0.0014,868 "reward": 4.2208333015441895,869 "reward_std": 0.23445481061935425,870 "rewards/accuracy_reward": 2.745833396911621,871 "rewards/format_reward": 1.0,872 "step": 54,873 "temporal_rewards": 1.0874 },875 {876 "all_correct": 1.0,877 "all_wrong": 0.0,878 "completion_length": 404.46875,879 "epoch": 0.09401709401709402,880 "grad_norm": 0.9479258360176128,881 "kl": 0.036468505859375,882 "learning_rate": 4.891741052068563e-07,883 "loss": 0.0015,884 "reward": 4.464583396911621,885 "reward_std": 0.05090419948101044,886 "rewards/accuracy_reward": 2.977083206176758,887 "rewards/format_reward": 1.0,888 "step": 55,889 "temporal_rewards": 1.0890 },891 {892 "all_correct": 1.0,893 "all_wrong": 0.0,894 "completion_length": 419.21875,895 "epoch": 0.09572649572649573,896 "grad_norm": 1.185262142708624,897 "kl": 0.0364990234375,898 "learning_rate": 4.887798554868387e-07,899 "loss": 0.0015,900 "reward": 4.202083110809326,901 "reward_std": 0.08526714146137238,902 "rewards/accuracy_reward": 2.7333333492279053,903 "rewards/format_reward": 1.0,904 "step": 56,905 "temporal_rewards": 1.0906 },907 {908 "all_correct": 1.0,909 "all_wrong": 0.0,910 "completion_length": 427.78125,911 "epoch": 0.09743589743589744,912 "grad_norm": 0.6149452604974177,913 "kl": 0.033416748046875,914 "learning_rate": 4.883787194871841e-07,915 "loss": 0.0013,916 "reward": 4.359375,917 "reward_std": 0.03125,918 "rewards/accuracy_reward": 2.871875047683716,919 "rewards/format_reward": 1.0,920 "step": 57,921 "temporal_rewards": 1.0922 },923 {924 "all_correct": 1.0,925 "all_wrong": 0.0,926 "completion_length": 396.4375,927 "epoch": 0.09914529914529914,928 "grad_norm": 1.1440282896265581,929 "kl": 0.039398193359375,930 "learning_rate": 4.879707087764336e-07,931 "loss": 0.0016,932 "reward": 4.335416793823242,933 "reward_std": 0.06786316633224487,934 "rewards/accuracy_reward": 2.835416793823242,935 "rewards/format_reward": 1.0,936 "step": 58,937 "temporal_rewards": 1.0938 },939 {940 "all_correct": 1.0,941 "all_wrong": 0.0,942 "completion_length": 403.34375,943 "epoch": 0.10085470085470086,944 "grad_norm": 1.1667079878189268,945 "kl": 0.0361328125,946 "learning_rate": 4.875558351213917e-07,947 "loss": 0.0014,948 "reward": 4.196874618530273,949 "reward_std": 0.07863453030586243,950 "rewards/accuracy_reward": 2.7093749046325684,951 "rewards/format_reward": 1.0,952 "step": 59,953 "temporal_rewards": 1.0954 },955 {956 "all_correct": 1.0,957 "all_wrong": 0.0,958 "completion_length": 402.3125,959 "epoch": 0.10256410256410256,960 "grad_norm": 0.7257762494267078,961 "kl": 0.04315185546875,962 "learning_rate": 4.871341104867864e-07,963 "loss": 0.0017,964 "reward": 4.106249809265137,965 "reward_std": 0.03750002384185791,966 "rewards/accuracy_reward": 2.606250047683716,967 "rewards/format_reward": 1.0,968 "step": 60,969 "temporal_rewards": 1.0970 },971 {972 "all_correct": 0.875,973 "all_wrong": 0.0,974 "completion_length": 390.375,975 "epoch": 0.10427350427350428,976 "grad_norm": 1.246730170179853,977 "kl": 0.038177490234375,978 "learning_rate": 4.86705547034924e-07,979 "loss": 0.0015,980 "reward": 4.362500190734863,981 "reward_std": 0.18642225861549377,982 "rewards/accuracy_reward": 2.9124999046325684,983 "rewards/format_reward": 1.0,984 "step": 61,985 "temporal_rewards": 0.875986 },987 {988 "all_correct": 1.0,989 "all_wrong": 0.0,990 "completion_length": 387.5625,991 "epoch": 0.10598290598290598,992 "grad_norm": 0.757303226225468,993 "kl": 0.0382080078125,994 "learning_rate": 4.862701571253386e-07,995 "loss": 0.0015,996 "reward": 4.3645830154418945,997 "reward_std": 0.020833373069763184,998 "rewards/accuracy_reward": 2.8645832538604736,999 "rewards/format_reward": 1.0,1000 "step": 62,1001 "temporal_rewards": 1.01002 },1003 {1004 "all_correct": 1.0,1005 "all_wrong": 0.0,1006 "completion_length": 403.84375,1007 "epoch": 0.1076923076923077,1008 "grad_norm": 0.7830222478805806,1009 "kl": 0.0361328125,1010 "learning_rate": 4.858279533144357e-07,1011 "loss": 0.0014,1012 "reward": 4.366666793823242,1013 "reward_std": 0.016666710376739502,1014 "rewards/accuracy_reward": 2.866666793823242,1015 "rewards/format_reward": 1.0,1016 "step": 63,1017 "temporal_rewards": 1.01018 },1019 {1020 "all_correct": 1.0,1021 "all_wrong": 0.0,1022 "completion_length": 385.0,1023 "epoch": 0.1094017094017094,1024 "grad_norm": 0.7856721207258216,1025 "kl": 0.037872314453125,1026 "learning_rate": 4.853789483551299e-07,1027 "loss": 0.0015,1028 "reward": 4.328125,1029 "reward_std": 0.07114279270172119,1030 "rewards/accuracy_reward": 2.828125,1031 "rewards/format_reward": 1.0,1032 "step": 64,1033 "temporal_rewards": 1.01034 },1035 {1036 "all_correct": 1.0,1037 "all_wrong": 0.0,1038 "completion_length": 391.6875,1039 "epoch": 0.1111111111111111,1040 "grad_norm": 0.5081050946401335,1041 "kl": 0.040435791015625,1042 "learning_rate": 4.849231551964771e-07,1043 "loss": 0.0016,1044 "reward": 4.213541507720947,1045 "reward_std": 0.059839192777872086,1046 "rewards/accuracy_reward": 2.7135415077209473,1047 "rewards/format_reward": 1.0,1048 "step": 65,1049 "temporal_rewards": 1.01050 },1051 {1052 "all_correct": 1.0,1053 "all_wrong": 0.0,1054 "completion_length": 390.4375,1055 "epoch": 0.11282051282051282,1056 "grad_norm": 0.7425251071608985,1057 "kl": 0.041046142578125,1058 "learning_rate": 4.844605869833011e-07,1059 "loss": 0.0016,1060 "reward": 4.2239580154418945,1061 "reward_std": 0.05208335071802139,1062 "rewards/accuracy_reward": 2.7239584922790527,1063 "rewards/format_reward": 1.0,1064 "step": 66,1065 "temporal_rewards": 1.01066 },1067 {1068 "all_correct": 1.0,1069 "all_wrong": 0.0,1070 "completion_length": 393.59375,1071 "epoch": 0.11452991452991453,1072 "grad_norm": 1.7762232281747519,1073 "kl": 0.03778076171875,1074 "learning_rate": 4.839912570558147e-07,1075 "loss": 0.0015,1076 "reward": 4.457291603088379,1077 "reward_std": 0.08541667461395264,1078 "rewards/accuracy_reward": 2.9635417461395264,1079 "rewards/format_reward": 1.0,1080 "step": 67,1081 "temporal_rewards": 1.01082 },1083 {1084 "all_correct": 1.0,1085 "all_wrong": 0.0,1086 "completion_length": 370.96875,1087 "epoch": 0.11623931623931624,1088 "grad_norm": 1.0171107252467595,1089 "kl": 0.039794921875,1090 "learning_rate": 4.835151789492348e-07,1091 "loss": 0.0016,1092 "reward": 4.327083587646484,1093 "reward_std": 0.06022907793521881,1094 "rewards/accuracy_reward": 2.8333334922790527,1095 "rewards/format_reward": 1.0,1096 "step": 68,1097 "temporal_rewards": 1.01098 },1099 {1100 "all_correct": 1.0,1101 "all_wrong": 0.0,1102 "completion_length": 372.0,1103 "epoch": 0.11794871794871795,1104 "grad_norm": 0.8418226397769828,1105 "kl": 0.038909912109375,1106 "learning_rate": 4.830323663933919e-07,1107 "loss": 0.0016,1108 "reward": 4.474999904632568,1109 "reward_std": 0.034150637686252594,1110 "rewards/accuracy_reward": 2.981250047683716,1111 "rewards/format_reward": 1.0,1112 "step": 69,1113 "temporal_rewards": 1.01114 },1115 {1116 "all_correct": 1.0,1117 "all_wrong": 0.0,1118 "completion_length": 358.40625,1119 "epoch": 0.11965811965811966,1120 "grad_norm": 1.2733531515130359,1121 "kl": 0.036468505859375,1122 "learning_rate": 4.825428333123346e-07,1123 "loss": 0.0015,1124 "reward": 4.4375,1125 "reward_std": 0.06786306202411652,1126 "rewards/accuracy_reward": 2.968750238418579,1127 "rewards/format_reward": 1.0,1128 "step": 70,1129 "temporal_rewards": 1.01130 },1131 {1132 "all_correct": 1.0,1133 "all_wrong": 0.0,1134 "completion_length": 381.875,1135 "epoch": 0.12136752136752137,1136 "grad_norm": 1.1124941217119864,1137 "kl": 0.039459228515625,1138 "learning_rate": 4.820465938239273e-07,1139 "loss": 0.0016,1140 "reward": 4.077083110809326,1141 "reward_std": 0.09583336114883423,1142 "rewards/accuracy_reward": 2.5833332538604736,1143 "rewards/format_reward": 1.0,1144 "step": 71,1145 "temporal_rewards": 1.01146 },1147 {1148 "all_correct": 1.0,1149 "all_wrong": 0.0,1150 "completion_length": 395.0625,1151 "epoch": 0.12307692307692308,1152 "grad_norm": 0.7611058988515808,1153 "kl": 0.037384033203125,1154 "learning_rate": 4.815436622394441e-07,1155 "loss": 0.0015,1156 "reward": 4.3333330154418945,1157 "reward_std": 0.07452815771102905,1158 "rewards/accuracy_reward": 2.8333330154418945,1159 "rewards/format_reward": 1.0,1160 "step": 72,1161 "temporal_rewards": 1.01162 },1163 {1164 "all_correct": 1.0,1165 "all_wrong": 0.0,1166 "completion_length": 401.3125,1167 "epoch": 0.12478632478632479,1168 "grad_norm": 0.7580960727925862,1169 "kl": 0.03826904296875,1170 "learning_rate": 4.810340530631549e-07,1171 "loss": 0.0015,1172 "reward": 4.196874618530273,1173 "reward_std": 0.042802631855010986,1174 "rewards/accuracy_reward": 2.7093749046325684,1175 "rewards/format_reward": 1.0,1176 "step": 73,1177 "temporal_rewards": 1.01178 },1179 {1180 "all_correct": 1.0,1181 "all_wrong": 0.0,1182 "completion_length": 396.8125,1183 "epoch": 0.1264957264957265,1184 "grad_norm": 1.0374945029632214,1185 "kl": 0.036468505859375,1186 "learning_rate": 4.805177809919081e-07,1187 "loss": 0.0015,1188 "reward": 4.40625,1189 "reward_std": 0.13188013434410095,1190 "rewards/accuracy_reward": 2.9124999046325684,1191 "rewards/format_reward": 1.0,1192 "step": 74,1193 "temporal_rewards": 1.01194 },1195 {1196 "all_correct": 1.0,1197 "all_wrong": 0.0,1198 "completion_length": 369.5,1199 "epoch": 0.1282051282051282,1200 "grad_norm": 1.4163910687642582,