| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 400, | |
| "global_step": 234, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004273504273504274, | |
| "grad_norm": 13.34490065414622, | |
| "learning_rate": 4.166666666666666e-08, | |
| "logits/chosen": -0.0994097888469696, | |
| "logits/rejected": -0.05551636964082718, | |
| "logps/chosen": -0.3415659964084625, | |
| "logps/rejected": -0.47305911779403687, | |
| "loss": 1.6161, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.8539150357246399, | |
| "rewards/margins": 0.32873278856277466, | |
| "rewards/rejected": -1.1826478242874146, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021367521367521368, | |
| "grad_norm": 13.589601768721613, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "logits/chosen": -0.1462363749742508, | |
| "logits/rejected": -0.14229728281497955, | |
| "logps/chosen": -0.4933474361896515, | |
| "logps/rejected": -0.5222159624099731, | |
| "loss": 1.6602, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -1.2333686351776123, | |
| "rewards/margins": 0.07217135280370712, | |
| "rewards/rejected": -1.3055399656295776, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.042735042735042736, | |
| "grad_norm": 8.87565340921273, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "logits/chosen": -0.06335792690515518, | |
| "logits/rejected": -0.02190782129764557, | |
| "logps/chosen": -0.42108288407325745, | |
| "logps/rejected": -0.4607582688331604, | |
| "loss": 1.6154, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -1.052707314491272, | |
| "rewards/margins": 0.09918837249279022, | |
| "rewards/rejected": -1.151895523071289, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0641025641025641, | |
| "grad_norm": 8.660979149745721, | |
| "learning_rate": 6.249999999999999e-07, | |
| "logits/chosen": -0.03177808225154877, | |
| "logits/rejected": -0.046066295355558395, | |
| "logps/chosen": -0.4381338953971863, | |
| "logps/rejected": -0.4405650198459625, | |
| "loss": 1.6068, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -1.095334768295288, | |
| "rewards/margins": 0.006077909376472235, | |
| "rewards/rejected": -1.1014125347137451, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08547008547008547, | |
| "grad_norm": 11.952248546593465, | |
| "learning_rate": 8.333333333333333e-07, | |
| "logits/chosen": -0.15588010847568512, | |
| "logits/rejected": -0.1342618316411972, | |
| "logps/chosen": -0.4571867883205414, | |
| "logps/rejected": -0.4777159094810486, | |
| "loss": 1.614, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.1429669857025146, | |
| "rewards/margins": 0.0513228178024292, | |
| "rewards/rejected": -1.1942898035049438, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10683760683760683, | |
| "grad_norm": 7.754135247761182, | |
| "learning_rate": 9.999440509051367e-07, | |
| "logits/chosen": -0.11927574872970581, | |
| "logits/rejected": -0.09945164620876312, | |
| "logps/chosen": -0.4351424276828766, | |
| "logps/rejected": -0.4545895457267761, | |
| "loss": 1.578, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -1.0878560543060303, | |
| "rewards/margins": 0.048617832362651825, | |
| "rewards/rejected": -1.1364738941192627, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1282051282051282, | |
| "grad_norm": 7.538242001814111, | |
| "learning_rate": 9.979871469976195e-07, | |
| "logits/chosen": -0.10869207233190536, | |
| "logits/rejected": -0.10381748527288437, | |
| "logps/chosen": -0.34100422263145447, | |
| "logps/rejected": -0.3819103240966797, | |
| "loss": 1.5753, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.8525105714797974, | |
| "rewards/margins": 0.10226528346538544, | |
| "rewards/rejected": -0.9547758102416992, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14957264957264957, | |
| "grad_norm": 12.195052142813049, | |
| "learning_rate": 9.932452969617607e-07, | |
| "logits/chosen": -0.1797598898410797, | |
| "logits/rejected": -0.08878383785486221, | |
| "logps/chosen": -0.39731845259666443, | |
| "logps/rejected": -0.4476427435874939, | |
| "loss": 1.5794, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.9932962656021118, | |
| "rewards/margins": 0.12581071257591248, | |
| "rewards/rejected": -1.1191068887710571, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.17094017094017094, | |
| "grad_norm": 9.951467872149747, | |
| "learning_rate": 9.857450191464337e-07, | |
| "logits/chosen": -0.1144944578409195, | |
| "logits/rejected": -0.09615515172481537, | |
| "logps/chosen": -0.4008924067020416, | |
| "logps/rejected": -0.4632578492164612, | |
| "loss": 1.5425, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.0022308826446533, | |
| "rewards/margins": 0.1559135764837265, | |
| "rewards/rejected": -1.158144474029541, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 9.233295836736795, | |
| "learning_rate": 9.755282581475767e-07, | |
| "logits/chosen": -0.13462205231189728, | |
| "logits/rejected": -0.11864318698644638, | |
| "logps/chosen": -0.38421568274497986, | |
| "logps/rejected": -0.4823899269104004, | |
| "loss": 1.563, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.9605391621589661, | |
| "rewards/margins": 0.24543562531471252, | |
| "rewards/rejected": -1.205974817276001, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.21367521367521367, | |
| "grad_norm": 8.910807590586288, | |
| "learning_rate": 9.626521502369983e-07, | |
| "logits/chosen": -0.19716130197048187, | |
| "logits/rejected": -0.15895314514636993, | |
| "logps/chosen": -0.37459030747413635, | |
| "logps/rejected": -0.4011107385158539, | |
| "loss": 1.5687, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.9364757537841797, | |
| "rewards/margins": 0.06630120426416397, | |
| "rewards/rejected": -1.002776861190796, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23504273504273504, | |
| "grad_norm": 8.741144982685217, | |
| "learning_rate": 9.471887038331684e-07, | |
| "logits/chosen": -0.07297103852033615, | |
| "logits/rejected": -0.045126646757125854, | |
| "logps/chosen": -0.3764435052871704, | |
| "logps/rejected": -0.4278687834739685, | |
| "loss": 1.5595, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.9411088228225708, | |
| "rewards/margins": 0.12856322526931763, | |
| "rewards/rejected": -1.0696719884872437, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.2564102564102564, | |
| "grad_norm": 9.65813041875177, | |
| "learning_rate": 9.29224396800933e-07, | |
| "logits/chosen": -0.00632941210642457, | |
| "logits/rejected": -0.04972558468580246, | |
| "logps/chosen": -0.3871188163757324, | |
| "logps/rejected": -0.42081761360168457, | |
| "loss": 1.5433, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.9677971005439758, | |
| "rewards/margins": 0.08424701541662216, | |
| "rewards/rejected": -1.052044153213501, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 9.22020584708003, | |
| "learning_rate": 9.088596928322157e-07, | |
| "logits/chosen": 0.052268654108047485, | |
| "logits/rejected": 0.07172085344791412, | |
| "logps/chosen": -0.5376943945884705, | |
| "logps/rejected": -0.6326268911361694, | |
| "loss": 1.5646, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.3442360162734985, | |
| "rewards/margins": 0.23733112215995789, | |
| "rewards/rejected": -1.5815670490264893, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.29914529914529914, | |
| "grad_norm": 13.413709386986495, | |
| "learning_rate": 8.862084796122997e-07, | |
| "logits/chosen": -0.014689329080283642, | |
| "logits/rejected": 0.023155853152275085, | |
| "logps/chosen": -0.4990696310997009, | |
| "logps/rejected": -0.5445122718811035, | |
| "loss": 1.5438, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -1.2476739883422852, | |
| "rewards/margins": 0.11360664665699005, | |
| "rewards/rejected": -1.3612806797027588, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.32051282051282054, | |
| "grad_norm": 11.427386040917398, | |
| "learning_rate": 8.613974319136957e-07, | |
| "logits/chosen": -0.13034021854400635, | |
| "logits/rejected": -0.12927956879138947, | |
| "logps/chosen": -0.41988080739974976, | |
| "logps/rejected": -0.5410174131393433, | |
| "loss": 1.5502, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.0497019290924072, | |
| "rewards/margins": 0.3028416335582733, | |
| "rewards/rejected": -1.352543592453003, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3418803418803419, | |
| "grad_norm": 8.049393765401897, | |
| "learning_rate": 8.34565303179429e-07, | |
| "logits/chosen": -0.028346195816993713, | |
| "logits/rejected": -0.03619036823511124, | |
| "logps/chosen": -0.45006662607192993, | |
| "logps/rejected": -0.5116376876831055, | |
| "loss": 1.5312, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.125166416168213, | |
| "rewards/margins": 0.15392769873142242, | |
| "rewards/rejected": -1.2790940999984741, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36324786324786323, | |
| "grad_norm": 10.027579862082366, | |
| "learning_rate": 8.058621495575031e-07, | |
| "logits/chosen": -0.14632529020309448, | |
| "logits/rejected": -0.09482773393392563, | |
| "logps/chosen": -0.4204605221748352, | |
| "logps/rejected": -0.5197293758392334, | |
| "loss": 1.5235, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.051151156425476, | |
| "rewards/margins": 0.24817219376564026, | |
| "rewards/rejected": -1.299323320388794, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 11.778153989300435, | |
| "learning_rate": 7.754484907260512e-07, | |
| "logits/chosen": -0.03493024781346321, | |
| "logits/rejected": -0.005648002028465271, | |
| "logps/chosen": -0.48695850372314453, | |
| "logps/rejected": -0.6510589718818665, | |
| "loss": 1.4999, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.2173962593078613, | |
| "rewards/margins": 0.4102511405944824, | |
| "rewards/rejected": -1.6276471614837646, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.405982905982906, | |
| "grad_norm": 9.302069462672513, | |
| "learning_rate": 7.434944122021836e-07, | |
| "logits/chosen": -0.03053552471101284, | |
| "logits/rejected": 0.008784343488514423, | |
| "logps/chosen": -0.48170527815818787, | |
| "logps/rejected": -0.7114425897598267, | |
| "loss": 1.4935, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.2042630910873413, | |
| "rewards/margins": 0.5743432641029358, | |
| "rewards/rejected": -1.7786064147949219, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.42735042735042733, | |
| "grad_norm": 15.155125489577449, | |
| "learning_rate": 7.101786141547828e-07, | |
| "logits/chosen": -0.04211338609457016, | |
| "logits/rejected": -0.033037807792425156, | |
| "logps/chosen": -0.5078177452087402, | |
| "logps/rejected": -0.5776672959327698, | |
| "loss": 1.515, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.2695444822311401, | |
| "rewards/margins": 0.17462393641471863, | |
| "rewards/rejected": -1.4441683292388916, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44871794871794873, | |
| "grad_norm": 16.110970936557738, | |
| "learning_rate": 6.756874120406714e-07, | |
| "logits/chosen": -0.09111454337835312, | |
| "logits/rejected": -0.05689894035458565, | |
| "logps/chosen": -0.53609699010849, | |
| "logps/rejected": -0.6224747896194458, | |
| "loss": 1.5115, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -1.3402423858642578, | |
| "rewards/margins": 0.21594436466693878, | |
| "rewards/rejected": -1.5561867952346802, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.4700854700854701, | |
| "grad_norm": 15.393371096801042, | |
| "learning_rate": 6.402136946530014e-07, | |
| "logits/chosen": -0.03849278762936592, | |
| "logits/rejected": 0.008776647970080376, | |
| "logps/chosen": -0.6499117016792297, | |
| "logps/rejected": -0.800051212310791, | |
| "loss": 1.5033, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.6247793436050415, | |
| "rewards/margins": 0.3753485679626465, | |
| "rewards/rejected": -2.0001277923583984, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.49145299145299143, | |
| "grad_norm": 9.808767922675623, | |
| "learning_rate": 6.039558454088795e-07, | |
| "logits/chosen": -0.03172523155808449, | |
| "logits/rejected": -0.04225381836295128, | |
| "logps/chosen": -0.5940315127372742, | |
| "logps/rejected": -0.7107864618301392, | |
| "loss": 1.5234, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.4850788116455078, | |
| "rewards/margins": 0.29188722372055054, | |
| "rewards/rejected": -1.7769660949707031, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 11.007927364769593, | |
| "learning_rate": 5.671166329088277e-07, | |
| "logits/chosen": -0.15589627623558044, | |
| "logits/rejected": -0.14328333735466003, | |
| "logps/chosen": -0.5461645126342773, | |
| "logps/rejected": -0.6048396825790405, | |
| "loss": 1.4873, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.3654112815856934, | |
| "rewards/margins": 0.14668798446655273, | |
| "rewards/rejected": -1.512099027633667, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5341880341880342, | |
| "grad_norm": 13.40581329613937, | |
| "learning_rate": 5.299020769725171e-07, | |
| "logits/chosen": -0.1017662063241005, | |
| "logits/rejected": -0.05507459491491318, | |
| "logps/chosen": -0.5289615392684937, | |
| "logps/rejected": -0.6923493146896362, | |
| "loss": 1.5008, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.322403907775879, | |
| "rewards/margins": 0.4084695875644684, | |
| "rewards/rejected": -1.730873465538025, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 10.393815491734465, | |
| "learning_rate": 4.925202964923683e-07, | |
| "logits/chosen": -0.01592491939663887, | |
| "logits/rejected": 0.10016925632953644, | |
| "logps/chosen": -0.545333981513977, | |
| "logps/rejected": -0.6946025490760803, | |
| "loss": 1.4829, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -1.3633348941802979, | |
| "rewards/margins": 0.3731713891029358, | |
| "rewards/rejected": -1.7365062236785889, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 12.796199500434565, | |
| "learning_rate": 4.5518034554828327e-07, | |
| "logits/chosen": -0.15349504351615906, | |
| "logits/rejected": -0.0872931182384491, | |
| "logps/chosen": -0.5930777788162231, | |
| "logps/rejected": -0.7436810731887817, | |
| "loss": 1.5284, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.4826946258544922, | |
| "rewards/margins": 0.3765079379081726, | |
| "rewards/rejected": -1.8592026233673096, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.5982905982905983, | |
| "grad_norm": 12.725430343830277, | |
| "learning_rate": 4.180910442924311e-07, | |
| "logits/chosen": -0.07290254533290863, | |
| "logits/rejected": -0.041871629655361176, | |
| "logps/chosen": -0.6430836915969849, | |
| "logps/rejected": -0.8283632397651672, | |
| "loss": 1.4923, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.607709288597107, | |
| "rewards/margins": 0.4631989598274231, | |
| "rewards/rejected": -2.070908308029175, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6196581196581197, | |
| "grad_norm": 11.626811297979486, | |
| "learning_rate": 3.814598111422513e-07, | |
| "logits/chosen": -0.11089099943637848, | |
| "logits/rejected": -0.08382478356361389, | |
| "logps/chosen": -0.6920641660690308, | |
| "logps/rejected": -0.7230005264282227, | |
| "loss": 1.5009, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -1.7301604747772217, | |
| "rewards/margins": 0.07734106481075287, | |
| "rewards/rejected": -1.8075014352798462, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 10.775765508549604, | |
| "learning_rate": 3.454915028125263e-07, | |
| "logits/chosen": -0.045754365622997284, | |
| "logits/rejected": -0.01993427611887455, | |
| "logps/chosen": -0.6500915288925171, | |
| "logps/rejected": -0.7409160733222961, | |
| "loss": 1.4998, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.6252286434173584, | |
| "rewards/margins": 0.22706131637096405, | |
| "rewards/rejected": -1.8522899150848389, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6623931623931624, | |
| "grad_norm": 14.344149642351613, | |
| "learning_rate": 3.1038726867353583e-07, | |
| "logits/chosen": -0.06552598625421524, | |
| "logits/rejected": -0.047451216727495193, | |
| "logps/chosen": -0.5619519948959351, | |
| "logps/rejected": -0.7450312972068787, | |
| "loss": 1.5001, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -1.4048798084259033, | |
| "rewards/margins": 0.4576982855796814, | |
| "rewards/rejected": -1.8625783920288086, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.6837606837606838, | |
| "grad_norm": 15.714808233921314, | |
| "learning_rate": 2.763434258421836e-07, | |
| "logits/chosen": -0.033187855035066605, | |
| "logits/rejected": 0.04786054790019989, | |
| "logps/chosen": -0.5716092586517334, | |
| "logps/rejected": -0.7472088932991028, | |
| "loss": 1.4866, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.4290231466293335, | |
| "rewards/margins": 0.4389989972114563, | |
| "rewards/rejected": -1.8680222034454346, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7051282051282052, | |
| "grad_norm": 12.685566519530033, | |
| "learning_rate": 2.4355036129704696e-07, | |
| "logits/chosen": 0.05670114606618881, | |
| "logits/rejected": 0.08839456737041473, | |
| "logps/chosen": -0.5581383109092712, | |
| "logps/rejected": -0.7727221250534058, | |
| "loss": 1.4701, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.395345687866211, | |
| "rewards/margins": 0.5364596247673035, | |
| "rewards/rejected": -1.9318053722381592, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.7264957264957265, | |
| "grad_norm": 15.872452807830859, | |
| "learning_rate": 2.121914671571633e-07, | |
| "logits/chosen": 0.03589984029531479, | |
| "logits/rejected": 0.07213737815618515, | |
| "logps/chosen": -0.5885176658630371, | |
| "logps/rejected": -0.7095993757247925, | |
| "loss": 1.4749, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.4712941646575928, | |
| "rewards/margins": 0.3027040958404541, | |
| "rewards/rejected": -1.773998498916626, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7478632478632479, | |
| "grad_norm": 10.782809870588457, | |
| "learning_rate": 1.824421150789106e-07, | |
| "logits/chosen": 0.08967064321041107, | |
| "logits/rejected": 0.07906897366046906, | |
| "logps/chosen": -0.7075825929641724, | |
| "logps/rejected": -0.8367185592651367, | |
| "loss": 1.4649, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.7689564228057861, | |
| "rewards/margins": 0.32283997535705566, | |
| "rewards/rejected": -2.091796398162842, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 19.45895258920461, | |
| "learning_rate": 1.5446867550656767e-07, | |
| "logits/chosen": 0.09190338104963303, | |
| "logits/rejected": 0.11659294366836548, | |
| "logps/chosen": -0.6805117130279541, | |
| "logps/rejected": -0.8247642517089844, | |
| "loss": 1.472, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -1.7012790441513062, | |
| "rewards/margins": 0.3606313467025757, | |
| "rewards/rejected": -2.061910629272461, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7905982905982906, | |
| "grad_norm": 12.591384251278834, | |
| "learning_rate": 1.284275872613028e-07, | |
| "logits/chosen": 0.09435538947582245, | |
| "logits/rejected": 0.0583331473171711, | |
| "logps/chosen": -0.6257596015930176, | |
| "logps/rejected": -0.7002542018890381, | |
| "loss": 1.4664, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.5643991231918335, | |
| "rewards/margins": 0.18623651564121246, | |
| "rewards/rejected": -1.7506357431411743, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.811965811965812, | |
| "grad_norm": 11.80568242770119, | |
| "learning_rate": 1.044644826718295e-07, | |
| "logits/chosen": 0.08234812319278717, | |
| "logits/rejected": 0.08470721542835236, | |
| "logps/chosen": -0.6872076988220215, | |
| "logps/rejected": -0.8852859735488892, | |
| "loss": 1.4845, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.7180191278457642, | |
| "rewards/margins": 0.4951957166194916, | |
| "rewards/rejected": -2.213214635848999, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 10.82000720224868, | |
| "learning_rate": 8.271337313934867e-08, | |
| "logits/chosen": 0.09420228004455566, | |
| "logits/rejected": 0.07983645796775818, | |
| "logps/chosen": -0.6325824856758118, | |
| "logps/rejected": -0.799116313457489, | |
| "loss": 1.4378, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.5814563035964966, | |
| "rewards/margins": 0.41633448004722595, | |
| "rewards/rejected": -1.9977906942367554, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.8547008547008547, | |
| "grad_norm": 10.37262835016834, | |
| "learning_rate": 6.329589969143517e-08, | |
| "logits/chosen": -0.02971937693655491, | |
| "logits/rejected": -0.009283095598220825, | |
| "logps/chosen": -0.6813124418258667, | |
| "logps/rejected": -0.7548044323921204, | |
| "loss": 1.4691, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.703281044960022, | |
| "rewards/margins": 0.18373003602027893, | |
| "rewards/rejected": -1.887010931968689, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8760683760683761, | |
| "grad_norm": 16.81161049195176, | |
| "learning_rate": 4.6320652716067555e-08, | |
| "logits/chosen": 0.057117123156785965, | |
| "logits/rejected": 0.033341288566589355, | |
| "logps/chosen": -0.5831505656242371, | |
| "logps/rejected": -0.7539544701576233, | |
| "loss": 1.4888, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -1.457876443862915, | |
| "rewards/margins": 0.4270097315311432, | |
| "rewards/rejected": -1.8848861455917358, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.8974358974358975, | |
| "grad_norm": 10.666184736871127, | |
| "learning_rate": 3.188256468013139e-08, | |
| "logits/chosen": 0.05842505767941475, | |
| "logits/rejected": 0.09373210370540619, | |
| "logps/chosen": -0.6092408299446106, | |
| "logps/rejected": -0.7288607954978943, | |
| "loss": 1.4636, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.523101806640625, | |
| "rewards/margins": 0.2990500330924988, | |
| "rewards/rejected": -1.8221518993377686, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9188034188034188, | |
| "grad_norm": 13.64833883057287, | |
| "learning_rate": 2.0062379228555525e-08, | |
| "logits/chosen": -0.03809903562068939, | |
| "logits/rejected": -0.026596253737807274, | |
| "logps/chosen": -0.6271125078201294, | |
| "logps/rejected": -0.6994240880012512, | |
| "loss": 1.4417, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -1.5677810907363892, | |
| "rewards/margins": 0.18077896535396576, | |
| "rewards/rejected": -1.7485601902008057, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.9401709401709402, | |
| "grad_norm": 11.81936042932564, | |
| "learning_rate": 1.0926199633097154e-08, | |
| "logits/chosen": 0.06771688163280487, | |
| "logits/rejected": 0.06887342035770416, | |
| "logps/chosen": -0.6525846123695374, | |
| "logps/rejected": -0.7262920141220093, | |
| "loss": 1.4542, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.631461501121521, | |
| "rewards/margins": 0.18426847457885742, | |
| "rewards/rejected": -1.815730094909668, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 13.440051766867484, | |
| "learning_rate": 4.5251191160326495e-09, | |
| "logits/chosen": -0.03224249184131622, | |
| "logits/rejected": 0.010183418169617653, | |
| "logps/chosen": -0.6054807901382446, | |
| "logps/rejected": -0.7111561894416809, | |
| "loss": 1.4562, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -1.513702154159546, | |
| "rewards/margins": 0.26418834924697876, | |
| "rewards/rejected": -1.7778904438018799, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.9829059829059829, | |
| "grad_norm": 11.45030232204693, | |
| "learning_rate": 8.949351161324225e-10, | |
| "logits/chosen": 0.15793287754058838, | |
| "logits/rejected": 0.19319342076778412, | |
| "logps/chosen": -0.8483369946479797, | |
| "logps/rejected": -0.9106278419494629, | |
| "loss": 1.4913, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.120842456817627, | |
| "rewards/margins": 0.1557270735502243, | |
| "rewards/rejected": -2.276569366455078, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 234, | |
| "total_flos": 0.0, | |
| "train_loss": 1.5168422256779468, | |
| "train_runtime": 9435.3453, | |
| "train_samples_per_second": 6.346, | |
| "train_steps_per_second": 0.025 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 234, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |