| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6523511823865181, | |
| "eval_steps": 50, | |
| "global_step": 900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007248346470961312, | |
| "grad_norm": 0.07235438376665115, | |
| "learning_rate": 4.998766400914329e-06, | |
| "logits/chosen": -1.9785633087158203, | |
| "logits/rejected": -2.5380234718322754, | |
| "logps/chosen": -0.28137442469596863, | |
| "logps/rejected": -0.37793582677841187, | |
| "loss": 7.3908, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.42206162214279175, | |
| "rewards/margins": 0.14484205842018127, | |
| "rewards/rejected": -0.5669037103652954, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014496692941922623, | |
| "grad_norm": 0.07455909997224808, | |
| "learning_rate": 4.9950668210706795e-06, | |
| "logits/chosen": -2.0576319694519043, | |
| "logits/rejected": -2.499069929122925, | |
| "logps/chosen": -0.2769497036933899, | |
| "logps/rejected": -0.33560773730278015, | |
| "loss": 7.3753, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.41542449593544006, | |
| "rewards/margins": 0.08798708021640778, | |
| "rewards/rejected": -0.503411591053009, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021745039412883936, | |
| "grad_norm": 0.09561269730329514, | |
| "learning_rate": 4.9889049115077e-06, | |
| "logits/chosen": -2.1018643379211426, | |
| "logits/rejected": -2.377673625946045, | |
| "logps/chosen": -0.2673099935054779, | |
| "logps/rejected": -0.3057115972042084, | |
| "loss": 7.4516, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.40096497535705566, | |
| "rewards/margins": 0.05760239437222481, | |
| "rewards/rejected": -0.45856744050979614, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.028993385883845247, | |
| "grad_norm": 0.08169445395469666, | |
| "learning_rate": 4.980286753286196e-06, | |
| "logits/chosen": -2.1627352237701416, | |
| "logits/rejected": -2.3873391151428223, | |
| "logps/chosen": -0.27636194229125977, | |
| "logps/rejected": -0.3703404664993286, | |
| "loss": 7.3888, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.4145428538322449, | |
| "rewards/margins": 0.14096775650978088, | |
| "rewards/rejected": -0.5555106401443481, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03624173235480656, | |
| "grad_norm": 0.08877147734165192, | |
| "learning_rate": 4.9692208514878445e-06, | |
| "logits/chosen": -2.1400368213653564, | |
| "logits/rejected": -2.44627046585083, | |
| "logps/chosen": -0.25384199619293213, | |
| "logps/rejected": -0.309225857257843, | |
| "loss": 7.518, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.3807629942893982, | |
| "rewards/margins": 0.08307582885026932, | |
| "rewards/rejected": -0.4638388752937317, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03624173235480656, | |
| "eval_logits/chosen": -2.1149837970733643, | |
| "eval_logits/rejected": -2.4759538173675537, | |
| "eval_logps/chosen": -0.28288090229034424, | |
| "eval_logps/rejected": -0.3434317111968994, | |
| "eval_loss": 0.9120966196060181, | |
| "eval_rewards/accuracies": 0.5089285969734192, | |
| "eval_rewards/chosen": -0.42432135343551636, | |
| "eval_rewards/margins": 0.09082622081041336, | |
| "eval_rewards/rejected": -0.5151475667953491, | |
| "eval_runtime": 30.2473, | |
| "eval_samples_per_second": 29.49, | |
| "eval_steps_per_second": 3.703, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04349007882576787, | |
| "grad_norm": 0.11774340271949768, | |
| "learning_rate": 4.9557181268217225e-06, | |
| "logits/chosen": -2.0567967891693115, | |
| "logits/rejected": -2.545668125152588, | |
| "logps/chosen": -0.2460743933916092, | |
| "logps/rejected": -0.32390663027763367, | |
| "loss": 7.3929, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.3691115975379944, | |
| "rewards/margins": 0.1167483925819397, | |
| "rewards/rejected": -0.4858599603176117, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05073842529672919, | |
| "grad_norm": 0.10012238472700119, | |
| "learning_rate": 4.939791904846869e-06, | |
| "logits/chosen": -1.9286606311798096, | |
| "logits/rejected": -2.4620895385742188, | |
| "logps/chosen": -0.25992274284362793, | |
| "logps/rejected": -0.3665553629398346, | |
| "loss": 7.1863, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.3898841440677643, | |
| "rewards/margins": 0.15994893014431, | |
| "rewards/rejected": -0.5498330593109131, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.057986771767690494, | |
| "grad_norm": 0.07387609034776688, | |
| "learning_rate": 4.921457902821578e-06, | |
| "logits/chosen": -2.0778262615203857, | |
| "logits/rejected": -2.4614813327789307, | |
| "logps/chosen": -0.24521782994270325, | |
| "logps/rejected": -0.3450419306755066, | |
| "loss": 7.3662, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.3678267002105713, | |
| "rewards/margins": 0.1497361809015274, | |
| "rewards/rejected": -0.5175628662109375, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0652351182386518, | |
| "grad_norm": 0.09718403220176697, | |
| "learning_rate": 4.900734214192358e-06, | |
| "logits/chosen": -1.998186707496643, | |
| "logits/rejected": -2.4382071495056152, | |
| "logps/chosen": -0.24080593883991241, | |
| "logps/rejected": -0.3280099332332611, | |
| "loss": 7.268, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.36120888590812683, | |
| "rewards/margins": 0.13080602884292603, | |
| "rewards/rejected": -0.49201488494873047, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07248346470961312, | |
| "grad_norm": 0.08059138059616089, | |
| "learning_rate": 4.8776412907378845e-06, | |
| "logits/chosen": -1.894728422164917, | |
| "logits/rejected": -2.475407838821411, | |
| "logps/chosen": -0.2019994705915451, | |
| "logps/rejected": -0.29733040928840637, | |
| "loss": 7.2544, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.30299919843673706, | |
| "rewards/margins": 0.14299637079238892, | |
| "rewards/rejected": -0.445995569229126, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07248346470961312, | |
| "eval_logits/chosen": -2.1233773231506348, | |
| "eval_logits/rejected": -2.4810428619384766, | |
| "eval_logps/chosen": -0.24400465190410614, | |
| "eval_logps/rejected": -0.3260345160961151, | |
| "eval_loss": 0.8920583724975586, | |
| "eval_rewards/accuracies": 0.5625, | |
| "eval_rewards/chosen": -0.3660070598125458, | |
| "eval_rewards/margins": 0.12304472178220749, | |
| "eval_rewards/rejected": -0.4890517592430115, | |
| "eval_runtime": 30.28, | |
| "eval_samples_per_second": 29.458, | |
| "eval_steps_per_second": 3.699, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07973181118057443, | |
| "grad_norm": 0.060006480664014816, | |
| "learning_rate": 4.852201922385564e-06, | |
| "logits/chosen": -2.1128883361816406, | |
| "logits/rejected": -2.5161118507385254, | |
| "logps/chosen": -0.23731942474842072, | |
| "logps/rejected": -0.31481316685676575, | |
| "loss": 7.1762, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3559790849685669, | |
| "rewards/margins": 0.11624068021774292, | |
| "rewards/rejected": -0.4722197651863098, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08698015765153574, | |
| "grad_norm": 0.0602690726518631, | |
| "learning_rate": 4.824441214720629e-06, | |
| "logits/chosen": -2.0212602615356445, | |
| "logits/rejected": -2.4894890785217285, | |
| "logps/chosen": -0.18956038355827332, | |
| "logps/rejected": -0.31935763359069824, | |
| "loss": 7.1442, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.2843405604362488, | |
| "rewards/margins": 0.19469590485095978, | |
| "rewards/rejected": -0.47903648018836975, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09422850412249706, | |
| "grad_norm": 0.06425223499536514, | |
| "learning_rate": 4.794386564209953e-06, | |
| "logits/chosen": -2.06717848777771, | |
| "logits/rejected": -2.5086405277252197, | |
| "logps/chosen": -0.2075866460800171, | |
| "logps/rejected": -0.3124083876609802, | |
| "loss": 7.1449, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.31137990951538086, | |
| "rewards/margins": 0.15723267197608948, | |
| "rewards/rejected": -0.46861258149147034, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.10147685059345837, | |
| "grad_norm": 0.08793757855892181, | |
| "learning_rate": 4.762067631165049e-06, | |
| "logits/chosen": -2.0242953300476074, | |
| "logits/rejected": -2.4943737983703613, | |
| "logps/chosen": -0.19989363849163055, | |
| "logps/rejected": -0.32057636976242065, | |
| "loss": 7.122, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.29984045028686523, | |
| "rewards/margins": 0.18102414906024933, | |
| "rewards/rejected": -0.48086461424827576, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.10872519706441967, | |
| "grad_norm": 0.09698841720819473, | |
| "learning_rate": 4.72751631047092e-06, | |
| "logits/chosen": -2.098658800125122, | |
| "logits/rejected": -2.582111120223999, | |
| "logps/chosen": -0.18897351622581482, | |
| "logps/rejected": -0.363709419965744, | |
| "loss": 7.1253, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.28346025943756104, | |
| "rewards/margins": 0.2621038556098938, | |
| "rewards/rejected": -0.5455641150474548, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.10872519706441967, | |
| "eval_logits/chosen": -2.1921768188476562, | |
| "eval_logits/rejected": -2.556370973587036, | |
| "eval_logps/chosen": -0.22214026749134064, | |
| "eval_logps/rejected": -0.3240993022918701, | |
| "eval_loss": 0.8764163851737976, | |
| "eval_rewards/accuracies": 0.5803571343421936, | |
| "eval_rewards/chosen": -0.33321040868759155, | |
| "eval_rewards/margins": 0.15293852984905243, | |
| "eval_rewards/rejected": -0.48614898324012756, | |
| "eval_runtime": 30.273, | |
| "eval_samples_per_second": 29.465, | |
| "eval_steps_per_second": 3.7, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11597354353538099, | |
| "grad_norm": 0.09119638800621033, | |
| "learning_rate": 4.690766700109659e-06, | |
| "logits/chosen": -2.2321066856384277, | |
| "logits/rejected": -2.5475690364837646, | |
| "logps/chosen": -0.19602122902870178, | |
| "logps/rejected": -0.27682799100875854, | |
| "loss": 7.0307, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.2940318286418915, | |
| "rewards/margins": 0.12121014297008514, | |
| "rewards/rejected": -0.4152420163154602, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1232218900063423, | |
| "grad_norm": 0.11496366560459137, | |
| "learning_rate": 4.65185506750986e-06, | |
| "logits/chosen": -2.1506271362304688, | |
| "logits/rejected": -2.506520986557007, | |
| "logps/chosen": -0.18018664419651031, | |
| "logps/rejected": -0.3084454834461212, | |
| "loss": 7.1652, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.27028003334999084, | |
| "rewards/margins": 0.19238826632499695, | |
| "rewards/rejected": -0.462668240070343, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.1304702364773036, | |
| "grad_norm": 0.09201692789793015, | |
| "learning_rate": 4.610819813755038e-06, | |
| "logits/chosen": -2.285245895385742, | |
| "logits/rejected": -2.573941707611084, | |
| "logps/chosen": -0.1844937801361084, | |
| "logps/rejected": -0.3112415075302124, | |
| "loss": 7.0022, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.2767406404018402, | |
| "rewards/margins": 0.1901216208934784, | |
| "rewards/rejected": -0.4668622612953186, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.13771858294826492, | |
| "grad_norm": 0.13191230595111847, | |
| "learning_rate": 4.567701435686405e-06, | |
| "logits/chosen": -2.265538454055786, | |
| "logits/rejected": -2.569638252258301, | |
| "logps/chosen": -0.21122264862060547, | |
| "logps/rejected": -0.3469196856021881, | |
| "loss": 7.004, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.3168340027332306, | |
| "rewards/margins": 0.20354552567005157, | |
| "rewards/rejected": -0.5203795433044434, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.14496692941922623, | |
| "grad_norm": 0.1503789722919464, | |
| "learning_rate": 4.522542485937369e-06, | |
| "logits/chosen": -2.2021026611328125, | |
| "logits/rejected": -2.655961513519287, | |
| "logps/chosen": -0.19664816558361053, | |
| "logps/rejected": -0.34679359197616577, | |
| "loss": 7.0539, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.2949722409248352, | |
| "rewards/margins": 0.22521813213825226, | |
| "rewards/rejected": -0.5201903581619263, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14496692941922623, | |
| "eval_logits/chosen": -2.325611114501953, | |
| "eval_logits/rejected": -2.715676784515381, | |
| "eval_logps/chosen": -0.2261900007724762, | |
| "eval_logps/rejected": -0.3570065200328827, | |
| "eval_loss": 0.8565592169761658, | |
| "eval_rewards/accuracies": 0.5714285969734192, | |
| "eval_rewards/chosen": -0.3392849862575531, | |
| "eval_rewards/margins": 0.19622473418712616, | |
| "eval_rewards/rejected": -0.5355097055435181, | |
| "eval_runtime": 30.2681, | |
| "eval_samples_per_second": 29.47, | |
| "eval_steps_per_second": 3.7, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15221527589018755, | |
| "grad_norm": 0.12589682638645172, | |
| "learning_rate": 4.475387530939226e-06, | |
| "logits/chosen": -2.24491810798645, | |
| "logits/rejected": -2.739323616027832, | |
| "logps/chosen": -0.19889305531978607, | |
| "logps/rejected": -0.32832199335098267, | |
| "loss": 6.9013, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.2983395457267761, | |
| "rewards/margins": 0.1941433995962143, | |
| "rewards/rejected": -0.492482990026474, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.15946362236114886, | |
| "grad_norm": 0.1417039930820465, | |
| "learning_rate": 4.426283106939474e-06, | |
| "logits/chosen": -2.278883695602417, | |
| "logits/rejected": -2.7108216285705566, | |
| "logps/chosen": -0.2170490026473999, | |
| "logps/rejected": -0.3783469796180725, | |
| "loss": 6.9366, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.32557350397109985, | |
| "rewards/margins": 0.2419470101594925, | |
| "rewards/rejected": -0.5675204992294312, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.16671196883211017, | |
| "grad_norm": 0.1472843438386917, | |
| "learning_rate": 4.3752776740761495e-06, | |
| "logits/chosen": -2.16560697555542, | |
| "logits/rejected": -2.7911431789398193, | |
| "logps/chosen": -0.20280452072620392, | |
| "logps/rejected": -0.3874067962169647, | |
| "loss": 6.8823, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.3042067885398865, | |
| "rewards/margins": 0.2769034504890442, | |
| "rewards/rejected": -0.5811101794242859, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.1739603153030715, | |
| "grad_norm": 0.15212437510490417, | |
| "learning_rate": 4.322421568553529e-06, | |
| "logits/chosen": -2.2929282188415527, | |
| "logits/rejected": -2.8649649620056152, | |
| "logps/chosen": -0.1841541975736618, | |
| "logps/rejected": -0.3838128447532654, | |
| "loss": 6.7139, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.2762312889099121, | |
| "rewards/margins": 0.29948797821998596, | |
| "rewards/rejected": -0.5757192373275757, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1812086617740328, | |
| "grad_norm": 0.22261729836463928, | |
| "learning_rate": 4.267766952966369e-06, | |
| "logits/chosen": -2.436549663543701, | |
| "logits/rejected": -2.827937126159668, | |
| "logps/chosen": -0.2322833091020584, | |
| "logps/rejected": -0.39587104320526123, | |
| "loss": 6.7737, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.3484249711036682, | |
| "rewards/margins": 0.24538159370422363, | |
| "rewards/rejected": -0.5938066244125366, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1812086617740328, | |
| "eval_logits/chosen": -2.4983742237091064, | |
| "eval_logits/rejected": -2.8988120555877686, | |
| "eval_logps/chosen": -0.269414484500885, | |
| "eval_logps/rejected": -0.45964303612709045, | |
| "eval_loss": 0.816527247428894, | |
| "eval_rewards/accuracies": 0.5714285969734192, | |
| "eval_rewards/chosen": -0.4041217863559723, | |
| "eval_rewards/margins": 0.2853427231311798, | |
| "eval_rewards/rejected": -0.6894644498825073, | |
| "eval_runtime": 30.2705, | |
| "eval_samples_per_second": 29.468, | |
| "eval_steps_per_second": 3.7, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.18845700824499412, | |
| "grad_norm": 0.25962531566619873, | |
| "learning_rate": 4.211367764821722e-06, | |
| "logits/chosen": -2.3911795616149902, | |
| "logits/rejected": -2.9030754566192627, | |
| "logps/chosen": -0.23631009459495544, | |
| "logps/rejected": -0.44218096137046814, | |
| "loss": 6.6574, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.35446515679359436, | |
| "rewards/margins": 0.3088063597679138, | |
| "rewards/rejected": -0.6632715463638306, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.19570535471595543, | |
| "grad_norm": 0.29614314436912537, | |
| "learning_rate": 4.15327966330913e-06, | |
| "logits/chosen": -2.39563250541687, | |
| "logits/rejected": -2.8926572799682617, | |
| "logps/chosen": -0.2558010220527649, | |
| "logps/rejected": -0.5495272278785706, | |
| "loss": 6.6017, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.38370150327682495, | |
| "rewards/margins": 0.4405893385410309, | |
| "rewards/rejected": -0.8242908716201782, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.20295370118691675, | |
| "grad_norm": 0.2586583197116852, | |
| "learning_rate": 4.093559974371725e-06, | |
| "logits/chosen": -2.5368762016296387, | |
| "logits/rejected": -2.8856003284454346, | |
| "logps/chosen": -0.2823846638202667, | |
| "logps/rejected": -0.4850040376186371, | |
| "loss": 6.3849, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.4235769808292389, | |
| "rewards/margins": 0.30392909049987793, | |
| "rewards/rejected": -0.7275060415267944, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.21020204765787803, | |
| "grad_norm": 0.3820374608039856, | |
| "learning_rate": 4.032267634132442e-06, | |
| "logits/chosen": -2.404680013656616, | |
| "logits/rejected": -2.8940536975860596, | |
| "logps/chosen": -0.3108128011226654, | |
| "logps/rejected": -0.7274529337882996, | |
| "loss": 6.0103, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -0.46621912717819214, | |
| "rewards/margins": 0.6249603033065796, | |
| "rewards/rejected": -1.0911794900894165, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.21745039412883935, | |
| "grad_norm": 0.572814404964447, | |
| "learning_rate": 3.969463130731183e-06, | |
| "logits/chosen": -2.3861801624298096, | |
| "logits/rejected": -2.8026037216186523, | |
| "logps/chosen": -0.40264564752578735, | |
| "logps/rejected": -0.8128012418746948, | |
| "loss": 5.9678, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.6039685010910034, | |
| "rewards/margins": 0.6152334809303284, | |
| "rewards/rejected": -1.219201922416687, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.21745039412883935, | |
| "eval_logits/chosen": -2.519517421722412, | |
| "eval_logits/rejected": -2.8339390754699707, | |
| "eval_logps/chosen": -0.4459304213523865, | |
| "eval_logps/rejected": -0.8040717840194702, | |
| "eval_loss": 0.7266466617584229, | |
| "eval_rewards/accuracies": 0.5625, | |
| "eval_rewards/chosen": -0.6688956618309021, | |
| "eval_rewards/margins": 0.5372119545936584, | |
| "eval_rewards/rejected": -1.206107497215271, | |
| "eval_runtime": 30.2692, | |
| "eval_samples_per_second": 29.469, | |
| "eval_steps_per_second": 3.7, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22469874059980066, | |
| "grad_norm": 0.3507815897464752, | |
| "learning_rate": 3.905208444630326e-06, | |
| "logits/chosen": -2.4855399131774902, | |
| "logits/rejected": -2.8341915607452393, | |
| "logps/chosen": -0.36750203371047974, | |
| "logps/rejected": -0.8827482461929321, | |
| "loss": 5.5511, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.5512530207633972, | |
| "rewards/margins": 0.7728692889213562, | |
| "rewards/rejected": -1.324122428894043, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.23194708707076198, | |
| "grad_norm": 0.34762975573539734, | |
| "learning_rate": 3.839566987447492e-06, | |
| "logits/chosen": -2.312408924102783, | |
| "logits/rejected": -2.7844693660736084, | |
| "logps/chosen": -0.465009868144989, | |
| "logps/rejected": -1.074065923690796, | |
| "loss": 5.5495, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.6975148320198059, | |
| "rewards/margins": 0.9135842323303223, | |
| "rewards/rejected": -1.6110990047454834, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.2391954335417233, | |
| "grad_norm": 0.5242094397544861, | |
| "learning_rate": 3.772603539375929e-06, | |
| "logits/chosen": -2.4517929553985596, | |
| "logits/rejected": -2.7228286266326904, | |
| "logps/chosen": -0.5588141679763794, | |
| "logps/rejected": -1.1725430488586426, | |
| "loss": 5.526, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.8382211923599243, | |
| "rewards/margins": 0.9205933809280396, | |
| "rewards/rejected": -1.7588145732879639, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.2464437800126846, | |
| "grad_norm": 0.7856224179267883, | |
| "learning_rate": 3.7043841852542884e-06, | |
| "logits/chosen": -2.5230085849761963, | |
| "logits/rejected": -2.8241894245147705, | |
| "logps/chosen": -0.527363657951355, | |
| "logps/rejected": -1.1461597681045532, | |
| "loss": 5.0865, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.7910455465316772, | |
| "rewards/margins": 0.9281940460205078, | |
| "rewards/rejected": -1.7192394733428955, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.2536921264836459, | |
| "grad_norm": 0.5080834031105042, | |
| "learning_rate": 3.634976249348867e-06, | |
| "logits/chosen": -2.414911985397339, | |
| "logits/rejected": -2.8537158966064453, | |
| "logps/chosen": -0.5747151374816895, | |
| "logps/rejected": -1.4558517932891846, | |
| "loss": 5.0036, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.862072765827179, | |
| "rewards/margins": 1.3217047452926636, | |
| "rewards/rejected": -2.1837775707244873, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.2536921264836459, | |
| "eval_logits/chosen": -2.647620677947998, | |
| "eval_logits/rejected": -2.910278558731079, | |
| "eval_logps/chosen": -0.6226872205734253, | |
| "eval_logps/rejected": -1.3186976909637451, | |
| "eval_loss": 0.6240565776824951, | |
| "eval_rewards/accuracies": 0.5982142686843872, | |
| "eval_rewards/chosen": -0.9340308308601379, | |
| "eval_rewards/margins": 1.0440157651901245, | |
| "eval_rewards/rejected": -1.9780464172363281, | |
| "eval_runtime": 30.2634, | |
| "eval_samples_per_second": 29.475, | |
| "eval_steps_per_second": 3.701, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.2609404729546072, | |
| "grad_norm": 0.5225710868835449, | |
| "learning_rate": 3.564448228912682e-06, | |
| "logits/chosen": -2.5971312522888184, | |
| "logits/rejected": -2.8506340980529785, | |
| "logps/chosen": -0.6313252449035645, | |
| "logps/rejected": -1.4177578687667847, | |
| "loss": 5.2602, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.9469879269599915, | |
| "rewards/margins": 1.179648995399475, | |
| "rewards/rejected": -2.1266369819641113, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.26818881942556855, | |
| "grad_norm": 0.44064539670944214, | |
| "learning_rate": 3.4928697265869516e-06, | |
| "logits/chosen": -2.645890712738037, | |
| "logits/rejected": -2.8477931022644043, | |
| "logps/chosen": -0.6448010206222534, | |
| "logps/rejected": -1.6615798473358154, | |
| "loss": 4.9912, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.9672015905380249, | |
| "rewards/margins": 1.5251682996749878, | |
| "rewards/rejected": -2.492370128631592, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.27543716589652983, | |
| "grad_norm": 0.6242617964744568, | |
| "learning_rate": 3.4203113817116955e-06, | |
| "logits/chosen": -2.685136318206787, | |
| "logits/rejected": -2.8105804920196533, | |
| "logps/chosen": -0.6792098879814148, | |
| "logps/rejected": -1.753614068031311, | |
| "loss": 4.7972, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.0188149213790894, | |
| "rewards/margins": 1.611606240272522, | |
| "rewards/rejected": -2.6304211616516113, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.2826855123674912, | |
| "grad_norm": 0.5383561253547668, | |
| "learning_rate": 3.346844800613229e-06, | |
| "logits/chosen": -2.6124706268310547, | |
| "logits/rejected": -2.8373632431030273, | |
| "logps/chosen": -0.6230086088180542, | |
| "logps/rejected": -2.398132801055908, | |
| "loss": 4.7275, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -0.9345127940177917, | |
| "rewards/margins": 2.662686824798584, | |
| "rewards/rejected": -3.5971992015838623, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.28993385883845246, | |
| "grad_norm": 0.6497470736503601, | |
| "learning_rate": 3.272542485937369e-06, | |
| "logits/chosen": -2.5315146446228027, | |
| "logits/rejected": -2.892939805984497, | |
| "logps/chosen": -0.7851268649101257, | |
| "logps/rejected": -2.1644442081451416, | |
| "loss": 4.5874, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.1776902675628662, | |
| "rewards/margins": 2.0689759254455566, | |
| "rewards/rejected": -3.246666431427002, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.28993385883845246, | |
| "eval_logits/chosen": -2.704826831817627, | |
| "eval_logits/rejected": -2.969177007675171, | |
| "eval_logps/chosen": -0.7823955416679382, | |
| "eval_logps/rejected": -2.0450375080108643, | |
| "eval_loss": 0.5390450358390808, | |
| "eval_rewards/accuracies": 0.5982142686843872, | |
| "eval_rewards/chosen": -1.1735934019088745, | |
| "eval_rewards/margins": 1.8939628601074219, | |
| "eval_rewards/rejected": -3.067556142807007, | |
| "eval_runtime": 30.267, | |
| "eval_samples_per_second": 29.471, | |
| "eval_steps_per_second": 3.7, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2971822053094138, | |
| "grad_norm": 0.5891414880752563, | |
| "learning_rate": 3.1974777650980737e-06, | |
| "logits/chosen": -2.6691222190856934, | |
| "logits/rejected": -2.9309957027435303, | |
| "logps/chosen": -0.7303667068481445, | |
| "logps/rejected": -2.1775012016296387, | |
| "loss": 4.3201, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -1.0955501794815063, | |
| "rewards/margins": 2.170701503753662, | |
| "rewards/rejected": -3.2662513256073, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.3044305517803751, | |
| "grad_norm": 1.0978150367736816, | |
| "learning_rate": 3.121724717912138e-06, | |
| "logits/chosen": -2.57599139213562, | |
| "logits/rejected": -2.839123249053955, | |
| "logps/chosen": -0.7009271383285522, | |
| "logps/rejected": -2.5759215354919434, | |
| "loss": 4.1763, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -1.0513907670974731, | |
| "rewards/margins": 2.8124916553497314, | |
| "rewards/rejected": -3.863882541656494, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.31167889825133643, | |
| "grad_norm": 0.5167409777641296, | |
| "learning_rate": 3.045358103491357e-06, | |
| "logits/chosen": -2.6339075565338135, | |
| "logits/rejected": -2.956376314163208, | |
| "logps/chosen": -0.84259432554245, | |
| "logps/rejected": -2.8900070190429688, | |
| "loss": 4.1528, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -1.2638914585113525, | |
| "rewards/margins": 3.0711190700531006, | |
| "rewards/rejected": -4.335010528564453, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.3189272447222977, | |
| "grad_norm": 0.5213866829872131, | |
| "learning_rate": 2.9684532864643123e-06, | |
| "logits/chosen": -2.469393014907837, | |
| "logits/rejected": -2.858123302459717, | |
| "logps/chosen": -0.9314866065979004, | |
| "logps/rejected": -3.1660075187683105, | |
| "loss": 3.7696, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -1.3972299098968506, | |
| "rewards/margins": 3.3517813682556152, | |
| "rewards/rejected": -4.749011516571045, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.32617559119325906, | |
| "grad_norm": 0.47010573744773865, | |
| "learning_rate": 2.8910861626005774e-06, | |
| "logits/chosen": -2.497013568878174, | |
| "logits/rejected": -2.8785929679870605, | |
| "logps/chosen": -0.7696805596351624, | |
| "logps/rejected": -3.087029457092285, | |
| "loss": 4.3625, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -1.1545207500457764, | |
| "rewards/margins": 3.4760234355926514, | |
| "rewards/rejected": -4.630544185638428, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.32617559119325906, | |
| "eval_logits/chosen": -2.6146204471588135, | |
| "eval_logits/rejected": -2.9200844764709473, | |
| "eval_logps/chosen": -0.8188837766647339, | |
| "eval_logps/rejected": -2.42936110496521, | |
| "eval_loss": 0.49255281686782837, | |
| "eval_rewards/accuracies": 0.6339285969734192, | |
| "eval_rewards/chosen": -1.2283254861831665, | |
| "eval_rewards/margins": 2.4157161712646484, | |
| "eval_rewards/rejected": -3.6440417766571045, | |
| "eval_runtime": 30.2604, | |
| "eval_samples_per_second": 29.478, | |
| "eval_steps_per_second": 3.701, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.33342393766422035, | |
| "grad_norm": 0.8721221685409546, | |
| "learning_rate": 2.813333083910761e-06, | |
| "logits/chosen": -2.593501567840576, | |
| "logits/rejected": -2.9277596473693848, | |
| "logps/chosen": -0.8970834016799927, | |
| "logps/rejected": -3.1408064365386963, | |
| "loss": 3.9344, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -1.3456249237060547, | |
| "rewards/margins": 3.3655846118927, | |
| "rewards/rejected": -4.711209774017334, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.34067228413518164, | |
| "grad_norm": 0.9014137983322144, | |
| "learning_rate": 2.7352707832962865e-06, | |
| "logits/chosen": -2.5280098915100098, | |
| "logits/rejected": -2.828508138656616, | |
| "logps/chosen": -0.9087240099906921, | |
| "logps/rejected": -2.842564105987549, | |
| "loss": 3.8492, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -1.3630859851837158, | |
| "rewards/margins": 2.9007601737976074, | |
| "rewards/rejected": -4.263846397399902, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.347920630606143, | |
| "grad_norm": 1.2404453754425049, | |
| "learning_rate": 2.6569762988232838e-06, | |
| "logits/chosen": -2.408717393875122, | |
| "logits/rejected": -2.7902934551239014, | |
| "logps/chosen": -1.1201987266540527, | |
| "logps/rejected": -3.879173994064331, | |
| "loss": 3.5822, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -1.6802982091903687, | |
| "rewards/margins": 4.138463020324707, | |
| "rewards/rejected": -5.818760871887207, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.35516897707710426, | |
| "grad_norm": 0.49635380506515503, | |
| "learning_rate": 2.578526897695321e-06, | |
| "logits/chosen": -2.422421455383301, | |
| "logits/rejected": -2.8097214698791504, | |
| "logps/chosen": -1.0756808519363403, | |
| "logps/rejected": -3.880105495452881, | |
| "loss": 3.8437, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -1.6135212182998657, | |
| "rewards/margins": 4.206637382507324, | |
| "rewards/rejected": -5.8201584815979, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.3624173235480656, | |
| "grad_norm": 0.5894991755485535, | |
| "learning_rate": 2.5e-06, | |
| "logits/chosen": -2.443977117538452, | |
| "logits/rejected": -2.734081983566284, | |
| "logps/chosen": -1.1097285747528076, | |
| "logps/rejected": -3.500363826751709, | |
| "loss": 3.8597, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -1.664592981338501, | |
| "rewards/margins": 3.5859527587890625, | |
| "rewards/rejected": -5.250545501708984, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3624173235480656, | |
| "eval_logits/chosen": -2.4630043506622314, | |
| "eval_logits/rejected": -2.7617809772491455, | |
| "eval_logps/chosen": -1.2134082317352295, | |
| "eval_logps/rejected": -3.330636739730835, | |
| "eval_loss": 0.45747870206832886, | |
| "eval_rewards/accuracies": 0.6517857313156128, | |
| "eval_rewards/chosen": -1.8201123476028442, | |
| "eval_rewards/margins": 3.1758430004119873, | |
| "eval_rewards/rejected": -4.995955467224121, | |
| "eval_runtime": 30.254, | |
| "eval_samples_per_second": 29.484, | |
| "eval_steps_per_second": 3.702, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3696656700190269, | |
| "grad_norm": 0.844010591506958, | |
| "learning_rate": 2.4214731023046795e-06, | |
| "logits/chosen": -2.3450961112976074, | |
| "logits/rejected": -2.7201991081237793, | |
| "logps/chosen": -1.1768486499786377, | |
| "logps/rejected": -4.009206295013428, | |
| "loss": 3.6558, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -1.765273094177246, | |
| "rewards/margins": 4.248536586761475, | |
| "rewards/rejected": -6.0138092041015625, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.37691401648998824, | |
| "grad_norm": 0.6226333379745483, | |
| "learning_rate": 2.3430237011767166e-06, | |
| "logits/chosen": -2.3935933113098145, | |
| "logits/rejected": -2.7934298515319824, | |
| "logps/chosen": -1.06581711769104, | |
| "logps/rejected": -3.6901473999023438, | |
| "loss": 3.8086, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -1.59872567653656, | |
| "rewards/margins": 3.936495304107666, | |
| "rewards/rejected": -5.535221099853516, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.3841623629609495, | |
| "grad_norm": 0.8458526134490967, | |
| "learning_rate": 2.2647292167037143e-06, | |
| "logits/chosen": -2.4400205612182617, | |
| "logits/rejected": -2.767524003982544, | |
| "logps/chosen": -1.2901079654693604, | |
| "logps/rejected": -4.540966033935547, | |
| "loss": 3.3647, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -1.935161828994751, | |
| "rewards/margins": 4.87628698348999, | |
| "rewards/rejected": -6.811448097229004, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.39141070943191086, | |
| "grad_norm": 1.2383215427398682, | |
| "learning_rate": 2.186666916089239e-06, | |
| "logits/chosen": -2.4235644340515137, | |
| "logits/rejected": -2.823796272277832, | |
| "logps/chosen": -1.3300189971923828, | |
| "logps/rejected": -4.055625915527344, | |
| "loss": 3.6972, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/chosen": -1.9950287342071533, | |
| "rewards/margins": 4.088409900665283, | |
| "rewards/rejected": -6.083438873291016, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.39865905590287215, | |
| "grad_norm": 1.0018137693405151, | |
| "learning_rate": 2.1089138373994226e-06, | |
| "logits/chosen": -2.498884677886963, | |
| "logits/rejected": -2.8074941635131836, | |
| "logps/chosen": -1.4506316184997559, | |
| "logps/rejected": -3.9293131828308105, | |
| "loss": 3.6202, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -2.175947427749634, | |
| "rewards/margins": 3.718022108078003, | |
| "rewards/rejected": -5.893969535827637, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.39865905590287215, | |
| "eval_logits/chosen": -2.491316556930542, | |
| "eval_logits/rejected": -2.792928457260132, | |
| "eval_logps/chosen": -1.7172259092330933, | |
| "eval_logps/rejected": -4.1004533767700195, | |
| "eval_loss": 0.43277421593666077, | |
| "eval_rewards/accuracies": 0.6875, | |
| "eval_rewards/chosen": -2.575838804244995, | |
| "eval_rewards/margins": 3.574841260910034, | |
| "eval_rewards/rejected": -6.1506805419921875, | |
| "eval_runtime": 30.2374, | |
| "eval_samples_per_second": 29.5, | |
| "eval_steps_per_second": 3.704, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.4059074023738335, | |
| "grad_norm": 0.903642475605011, | |
| "learning_rate": 2.031546713535688e-06, | |
| "logits/chosen": -2.452147960662842, | |
| "logits/rejected": -2.799231767654419, | |
| "logps/chosen": -1.7636197805404663, | |
| "logps/rejected": -4.704730033874512, | |
| "loss": 3.2952, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -2.645429849624634, | |
| "rewards/margins": 4.411664962768555, | |
| "rewards/rejected": -7.057094573974609, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.4131557488447948, | |
| "grad_norm": 1.289736032485962, | |
| "learning_rate": 1.9546418965086444e-06, | |
| "logits/chosen": -2.556669235229492, | |
| "logits/rejected": -2.833976984024048, | |
| "logps/chosen": -1.901815414428711, | |
| "logps/rejected": -4.644237518310547, | |
| "loss": 3.3085, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -2.852722644805908, | |
| "rewards/margins": 4.113633632659912, | |
| "rewards/rejected": -6.9663567543029785, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.42040409531575607, | |
| "grad_norm": 1.4252225160598755, | |
| "learning_rate": 1.8782752820878636e-06, | |
| "logits/chosen": -2.364348888397217, | |
| "logits/rejected": -2.9077506065368652, | |
| "logps/chosen": -2.175013780593872, | |
| "logps/rejected": -6.0506911277771, | |
| "loss": 3.2304, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -3.2625205516815186, | |
| "rewards/margins": 5.813515663146973, | |
| "rewards/rejected": -9.07603645324707, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.4276524417867174, | |
| "grad_norm": 1.1662834882736206, | |
| "learning_rate": 1.8025222349019273e-06, | |
| "logits/chosen": -2.431143045425415, | |
| "logits/rejected": -2.8574650287628174, | |
| "logps/chosen": -2.1578497886657715, | |
| "logps/rejected": -5.375465393066406, | |
| "loss": 3.2736, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -3.2367749214172363, | |
| "rewards/margins": 4.826422691345215, | |
| "rewards/rejected": -8.063197135925293, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.4349007882576787, | |
| "grad_norm": 1.6105389595031738, | |
| "learning_rate": 1.7274575140626318e-06, | |
| "logits/chosen": -2.3821804523468018, | |
| "logits/rejected": -2.8510732650756836, | |
| "logps/chosen": -2.297203540802002, | |
| "logps/rejected": -5.809553146362305, | |
| "loss": 3.343, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -3.445805072784424, | |
| "rewards/margins": 5.268526077270508, | |
| "rewards/rejected": -8.714330673217773, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4349007882576787, | |
| "eval_logits/chosen": -2.515763521194458, | |
| "eval_logits/rejected": -2.8498146533966064, | |
| "eval_logps/chosen": -2.3867976665496826, | |
| "eval_logps/rejected": -5.036753177642822, | |
| "eval_loss": 0.3846723139286041, | |
| "eval_rewards/accuracies": 0.75, | |
| "eval_rewards/chosen": -3.5801963806152344, | |
| "eval_rewards/margins": 3.974933385848999, | |
| "eval_rewards/rejected": -7.5551300048828125, | |
| "eval_runtime": 30.235, | |
| "eval_samples_per_second": 29.502, | |
| "eval_steps_per_second": 3.704, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44214913472864004, | |
| "grad_norm": 1.5437382459640503, | |
| "learning_rate": 1.6531551993867717e-06, | |
| "logits/chosen": -2.5224862098693848, | |
| "logits/rejected": -2.862809896469116, | |
| "logps/chosen": -2.504312753677368, | |
| "logps/rejected": -5.347531318664551, | |
| "loss": 3.0472, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -3.7564690113067627, | |
| "rewards/margins": 4.264827728271484, | |
| "rewards/rejected": -8.021296501159668, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.4493974811996013, | |
| "grad_norm": 1.5009058713912964, | |
| "learning_rate": 1.5796886182883053e-06, | |
| "logits/chosen": -2.4633076190948486, | |
| "logits/rejected": -2.9186339378356934, | |
| "logps/chosen": -2.854260206222534, | |
| "logps/rejected": -6.636805057525635, | |
| "loss": 3.1013, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -4.281390190124512, | |
| "rewards/margins": 5.673816680908203, | |
| "rewards/rejected": -9.955206871032715, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.45664582767056267, | |
| "grad_norm": 1.2582628726959229, | |
| "learning_rate": 1.5071302734130488e-06, | |
| "logits/chosen": -2.5565848350524902, | |
| "logits/rejected": -2.8856873512268066, | |
| "logps/chosen": -2.7753074169158936, | |
| "logps/rejected": -5.89193058013916, | |
| "loss": 2.7849, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -4.162961483001709, | |
| "rewards/margins": 4.674934387207031, | |
| "rewards/rejected": -8.837895393371582, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.46389417414152395, | |
| "grad_norm": 1.6643486022949219, | |
| "learning_rate": 1.4355517710873184e-06, | |
| "logits/chosen": -2.4163150787353516, | |
| "logits/rejected": -2.8351335525512695, | |
| "logps/chosen": -2.870678663253784, | |
| "logps/rejected": -6.636674404144287, | |
| "loss": 2.6839, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -4.306017875671387, | |
| "rewards/margins": 5.648994445800781, | |
| "rewards/rejected": -9.955012321472168, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.4711425206124853, | |
| "grad_norm": 0.9412183165550232, | |
| "learning_rate": 1.3650237506511333e-06, | |
| "logits/chosen": -2.511819362640381, | |
| "logits/rejected": -2.9430336952209473, | |
| "logps/chosen": -3.0111937522888184, | |
| "logps/rejected": -6.41957950592041, | |
| "loss": 2.884, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -4.516791343688965, | |
| "rewards/margins": 5.11257791519165, | |
| "rewards/rejected": -9.62936782836914, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.4711425206124853, | |
| "eval_logits/chosen": -2.5462071895599365, | |
| "eval_logits/rejected": -2.890479326248169, | |
| "eval_logps/chosen": -3.189256429672241, | |
| "eval_logps/rejected": -6.0319504737854, | |
| "eval_loss": 0.35537806153297424, | |
| "eval_rewards/accuracies": 0.8303571343421936, | |
| "eval_rewards/chosen": -4.7838850021362305, | |
| "eval_rewards/margins": 4.264040470123291, | |
| "eval_rewards/rejected": -9.04792594909668, | |
| "eval_runtime": 30.2531, | |
| "eval_samples_per_second": 29.485, | |
| "eval_steps_per_second": 3.702, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.4783908670834466, | |
| "grad_norm": 2.0148541927337646, | |
| "learning_rate": 1.2956158147457116e-06, | |
| "logits/chosen": -2.5816783905029297, | |
| "logits/rejected": -2.832157850265503, | |
| "logps/chosen": -2.9875688552856445, | |
| "logps/rejected": -6.420091152191162, | |
| "loss": 2.7163, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -4.481352806091309, | |
| "rewards/margins": 5.14878511428833, | |
| "rewards/rejected": -9.630138397216797, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.4856392135544079, | |
| "grad_norm": 2.247664451599121, | |
| "learning_rate": 1.2273964606240718e-06, | |
| "logits/chosen": -2.524240255355835, | |
| "logits/rejected": -2.936284065246582, | |
| "logps/chosen": -3.2474262714385986, | |
| "logps/rejected": -6.681307315826416, | |
| "loss": 2.9501, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -4.871139049530029, | |
| "rewards/margins": 5.150821685791016, | |
| "rewards/rejected": -10.021961212158203, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.4928875600253692, | |
| "grad_norm": 2.2333178520202637, | |
| "learning_rate": 1.160433012552508e-06, | |
| "logits/chosen": -2.5046608448028564, | |
| "logits/rejected": -2.889239549636841, | |
| "logps/chosen": -3.5039947032928467, | |
| "logps/rejected": -6.702293395996094, | |
| "loss": 2.777, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -5.2559919357299805, | |
| "rewards/margins": 4.797448635101318, | |
| "rewards/rejected": -10.053441047668457, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.5001359064963306, | |
| "grad_norm": 2.373650074005127, | |
| "learning_rate": 1.0947915553696742e-06, | |
| "logits/chosen": -2.4730420112609863, | |
| "logits/rejected": -2.9917922019958496, | |
| "logps/chosen": -3.579982042312622, | |
| "logps/rejected": -7.557694435119629, | |
| "loss": 2.673, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -5.369973182678223, | |
| "rewards/margins": 5.9665679931640625, | |
| "rewards/rejected": -11.336541175842285, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.5073842529672918, | |
| "grad_norm": 5.8343729972839355, | |
| "learning_rate": 1.0305368692688175e-06, | |
| "logits/chosen": -2.479050397872925, | |
| "logits/rejected": -2.9097132682800293, | |
| "logps/chosen": -3.8712658882141113, | |
| "logps/rejected": -7.749911308288574, | |
| "loss": 2.6632, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -5.806898593902588, | |
| "rewards/margins": 5.817968845367432, | |
| "rewards/rejected": -11.62486743927002, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5073842529672918, | |
| "eval_logits/chosen": -2.528568744659424, | |
| "eval_logits/rejected": -2.898803949356079, | |
| "eval_logps/chosen": -3.8546855449676514, | |
| "eval_logps/rejected": -6.8783745765686035, | |
| "eval_loss": 0.32792916893959045, | |
| "eval_rewards/accuracies": 0.8482142686843872, | |
| "eval_rewards/chosen": -5.782027721405029, | |
| "eval_rewards/margins": 4.535533428192139, | |
| "eval_rewards/rejected": -10.317561149597168, | |
| "eval_runtime": 30.2435, | |
| "eval_samples_per_second": 29.494, | |
| "eval_steps_per_second": 3.703, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5146325994382531, | |
| "grad_norm": 2.118319034576416, | |
| "learning_rate": 9.677323658675594e-07, | |
| "logits/chosen": -2.5124125480651855, | |
| "logits/rejected": -2.872951030731201, | |
| "logps/chosen": -3.9242005348205566, | |
| "logps/rejected": -7.064782619476318, | |
| "loss": 2.6968, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -5.886300086975098, | |
| "rewards/margins": 4.710873603820801, | |
| "rewards/rejected": -10.597173690795898, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.5218809459092144, | |
| "grad_norm": 2.57857346534729, | |
| "learning_rate": 9.064400256282757e-07, | |
| "logits/chosen": -2.442990779876709, | |
| "logits/rejected": -2.9133260250091553, | |
| "logps/chosen": -4.32869291305542, | |
| "logps/rejected": -7.873679161071777, | |
| "loss": 2.6929, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -6.493039608001709, | |
| "rewards/margins": 5.317479610443115, | |
| "rewards/rejected": -11.810518264770508, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.5291292923801758, | |
| "grad_norm": 3.1820855140686035, | |
| "learning_rate": 8.467203366908708e-07, | |
| "logits/chosen": -2.4702260494232178, | |
| "logits/rejected": -2.948544979095459, | |
| "logps/chosen": -4.033299922943115, | |
| "logps/rejected": -7.3781843185424805, | |
| "loss": 2.6089, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -6.0499491691589355, | |
| "rewards/margins": 5.017327785491943, | |
| "rewards/rejected": -11.067277908325195, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.5363776388511371, | |
| "grad_norm": 1.360353708267212, | |
| "learning_rate": 7.886322351782782e-07, | |
| "logits/chosen": -2.4439663887023926, | |
| "logits/rejected": -2.93951678276062, | |
| "logps/chosen": -3.9287734031677246, | |
| "logps/rejected": -8.031271934509277, | |
| "loss": 2.3252, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -5.893160820007324, | |
| "rewards/margins": 6.15374755859375, | |
| "rewards/rejected": -12.046907424926758, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.5436259853220984, | |
| "grad_norm": 2.4719009399414062, | |
| "learning_rate": 7.322330470336314e-07, | |
| "logits/chosen": -2.4842166900634766, | |
| "logits/rejected": -2.930525302886963, | |
| "logps/chosen": -4.090015888214111, | |
| "logps/rejected": -7.851285457611084, | |
| "loss": 2.4229, | |
| "rewards/accuracies": 0.9624999761581421, | |
| "rewards/chosen": -6.135023593902588, | |
| "rewards/margins": 5.641904830932617, | |
| "rewards/rejected": -11.776927947998047, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.5436259853220984, | |
| "eval_logits/chosen": -2.5460832118988037, | |
| "eval_logits/rejected": -2.9118590354919434, | |
| "eval_logps/chosen": -4.3249192237854, | |
| "eval_logps/rejected": -7.5089521408081055, | |
| "eval_loss": 0.30500930547714233, | |
| "eval_rewards/accuracies": 0.8571428656578064, | |
| "eval_rewards/chosen": -6.4873785972595215, | |
| "eval_rewards/margins": 4.7760491371154785, | |
| "eval_rewards/rejected": -11.263427734375, | |
| "eval_runtime": 30.2309, | |
| "eval_samples_per_second": 29.506, | |
| "eval_steps_per_second": 3.705, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.5508743317930597, | |
| "grad_norm": 1.5037100315093994, | |
| "learning_rate": 6.775784314464717e-07, | |
| "logits/chosen": -2.536068916320801, | |
| "logits/rejected": -2.9130196571350098, | |
| "logps/chosen": -4.38153076171875, | |
| "logps/rejected": -7.925049781799316, | |
| "loss": 2.2663, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -6.572295188903809, | |
| "rewards/margins": 5.315279960632324, | |
| "rewards/rejected": -11.887575149536133, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.5581226782640211, | |
| "grad_norm": 2.590315818786621, | |
| "learning_rate": 6.247223259238511e-07, | |
| "logits/chosen": -2.5102131366729736, | |
| "logits/rejected": -2.9389328956604004, | |
| "logps/chosen": -4.440621852874756, | |
| "logps/rejected": -7.853024482727051, | |
| "loss": 2.5056, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -6.660933017730713, | |
| "rewards/margins": 5.118602752685547, | |
| "rewards/rejected": -11.779535293579102, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.5653710247349824, | |
| "grad_norm": 1.7628798484802246, | |
| "learning_rate": 5.737168930605272e-07, | |
| "logits/chosen": -2.5515689849853516, | |
| "logits/rejected": -2.930510997772217, | |
| "logps/chosen": -4.128845691680908, | |
| "logps/rejected": -7.9085845947265625, | |
| "loss": 2.2382, | |
| "rewards/accuracies": 0.949999988079071, | |
| "rewards/chosen": -6.193268775939941, | |
| "rewards/margins": 5.669609069824219, | |
| "rewards/rejected": -11.862876892089844, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.5726193712059436, | |
| "grad_norm": 3.5818936824798584, | |
| "learning_rate": 5.24612469060774e-07, | |
| "logits/chosen": -2.4889330863952637, | |
| "logits/rejected": -2.87908935546875, | |
| "logps/chosen": -4.441933631896973, | |
| "logps/rejected": -8.435145378112793, | |
| "loss": 2.2271, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -6.662900447845459, | |
| "rewards/margins": 5.9898176193237305, | |
| "rewards/rejected": -12.652718544006348, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.5798677176769049, | |
| "grad_norm": 4.387265682220459, | |
| "learning_rate": 4.774575140626317e-07, | |
| "logits/chosen": -2.434931516647339, | |
| "logits/rejected": -2.9189987182617188, | |
| "logps/chosen": -4.684146404266357, | |
| "logps/rejected": -8.587119102478027, | |
| "loss": 2.2777, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -7.026219844818115, | |
| "rewards/margins": 5.854458808898926, | |
| "rewards/rejected": -12.880678176879883, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5798677176769049, | |
| "eval_logits/chosen": -2.5454742908477783, | |
| "eval_logits/rejected": -2.9115586280822754, | |
| "eval_logps/chosen": -4.826877117156982, | |
| "eval_logps/rejected": -8.15951919555664, | |
| "eval_loss": 0.2918218672275543, | |
| "eval_rewards/accuracies": 0.875, | |
| "eval_rewards/chosen": -7.2403154373168945, | |
| "eval_rewards/margins": 4.998963356018066, | |
| "eval_rewards/rejected": -12.239279747009277, | |
| "eval_runtime": 30.2636, | |
| "eval_samples_per_second": 29.474, | |
| "eval_steps_per_second": 3.701, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5871160641478663, | |
| "grad_norm": 3.7356505393981934, | |
| "learning_rate": 4.3229856431359516e-07, | |
| "logits/chosen": -2.456073522567749, | |
| "logits/rejected": -2.893204927444458, | |
| "logps/chosen": -4.553319931030273, | |
| "logps/rejected": -8.365021705627441, | |
| "loss": 2.5536, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -6.829980373382568, | |
| "rewards/margins": 5.717551231384277, | |
| "rewards/rejected": -12.547531127929688, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.5943644106188276, | |
| "grad_norm": 2.5566742420196533, | |
| "learning_rate": 3.891801862449629e-07, | |
| "logits/chosen": -2.439631938934326, | |
| "logits/rejected": -2.931784152984619, | |
| "logps/chosen": -4.470883846282959, | |
| "logps/rejected": -8.863119125366211, | |
| "loss": 2.279, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -6.706325531005859, | |
| "rewards/margins": 6.588353633880615, | |
| "rewards/rejected": -13.294679641723633, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.6016127570897889, | |
| "grad_norm": 3.1389355659484863, | |
| "learning_rate": 3.481449324901412e-07, | |
| "logits/chosen": -2.4623847007751465, | |
| "logits/rejected": -2.9169857501983643, | |
| "logps/chosen": -4.818366050720215, | |
| "logps/rejected": -8.847421646118164, | |
| "loss": 2.1291, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -7.227549076080322, | |
| "rewards/margins": 6.043582439422607, | |
| "rewards/rejected": -13.271130561828613, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.6088611035607502, | |
| "grad_norm": 2.467841625213623, | |
| "learning_rate": 3.092332998903416e-07, | |
| "logits/chosen": -2.5648622512817383, | |
| "logits/rejected": -2.8750557899475098, | |
| "logps/chosen": -4.850269794464111, | |
| "logps/rejected": -8.536476135253906, | |
| "loss": 2.2548, | |
| "rewards/accuracies": 0.949999988079071, | |
| "rewards/chosen": -7.275404930114746, | |
| "rewards/margins": 5.529307842254639, | |
| "rewards/rejected": -12.804713249206543, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.6161094500317115, | |
| "grad_norm": 2.4988455772399902, | |
| "learning_rate": 2.7248368952908055e-07, | |
| "logits/chosen": -2.439044952392578, | |
| "logits/rejected": -2.8845443725585938, | |
| "logps/chosen": -4.711085319519043, | |
| "logps/rejected": -8.678643226623535, | |
| "loss": 2.2852, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -7.066628456115723, | |
| "rewards/margins": 5.951335906982422, | |
| "rewards/rejected": -13.017964363098145, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.6161094500317115, | |
| "eval_logits/chosen": -2.5347232818603516, | |
| "eval_logits/rejected": -2.8995745182037354, | |
| "eval_logps/chosen": -4.948782444000244, | |
| "eval_logps/rejected": -8.324250221252441, | |
| "eval_loss": 0.2818227708339691, | |
| "eval_rewards/accuracies": 0.875, | |
| "eval_rewards/chosen": -7.423173427581787, | |
| "eval_rewards/margins": 5.063201427459717, | |
| "eval_rewards/rejected": -12.48637580871582, | |
| "eval_runtime": 30.2383, | |
| "eval_samples_per_second": 29.499, | |
| "eval_steps_per_second": 3.704, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.6233577965026729, | |
| "grad_norm": 3.6566624641418457, | |
| "learning_rate": 2.3793236883495164e-07, | |
| "logits/chosen": -2.522355556488037, | |
| "logits/rejected": -2.9092490673065186, | |
| "logps/chosen": -5.117705345153809, | |
| "logps/rejected": -8.789793014526367, | |
| "loss": 2.2219, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -7.676558017730713, | |
| "rewards/margins": 5.508131504058838, | |
| "rewards/rejected": -13.18468952178955, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.6306061429736342, | |
| "grad_norm": 3.4471309185028076, | |
| "learning_rate": 2.0561343579004716e-07, | |
| "logits/chosen": -2.478935480117798, | |
| "logits/rejected": -2.8984580039978027, | |
| "logps/chosen": -4.6364336013793945, | |
| "logps/rejected": -8.75157356262207, | |
| "loss": 2.0656, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -6.954649925231934, | |
| "rewards/margins": 6.1727094650268555, | |
| "rewards/rejected": -13.127359390258789, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.6378544894445954, | |
| "grad_norm": 5.826159477233887, | |
| "learning_rate": 1.7555878527937164e-07, | |
| "logits/chosen": -2.543975830078125, | |
| "logits/rejected": -2.8908252716064453, | |
| "logps/chosen": -4.740494728088379, | |
| "logps/rejected": -8.813726425170898, | |
| "loss": 2.2936, | |
| "rewards/accuracies": 0.9750000238418579, | |
| "rewards/chosen": -7.110742092132568, | |
| "rewards/margins": 6.1098480224609375, | |
| "rewards/rejected": -13.220590591430664, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.6451028359155567, | |
| "grad_norm": 1.7311068773269653, | |
| "learning_rate": 1.4779807761443638e-07, | |
| "logits/chosen": -2.4625258445739746, | |
| "logits/rejected": -2.8965325355529785, | |
| "logps/chosen": -5.210020542144775, | |
| "logps/rejected": -9.470269203186035, | |
| "loss": 2.1442, | |
| "rewards/accuracies": 0.9624999761581421, | |
| "rewards/chosen": -7.815030097961426, | |
| "rewards/margins": 6.3903727531433105, | |
| "rewards/rejected": -14.205403327941895, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.6523511823865181, | |
| "grad_norm": 3.3928353786468506, | |
| "learning_rate": 1.223587092621162e-07, | |
| "logits/chosen": -2.5978572368621826, | |
| "logits/rejected": -2.910475492477417, | |
| "logps/chosen": -5.280083656311035, | |
| "logps/rejected": -8.648152351379395, | |
| "loss": 2.6127, | |
| "rewards/accuracies": 0.9624999761581421, | |
| "rewards/chosen": -7.9201250076293945, | |
| "rewards/margins": 5.052102088928223, | |
| "rewards/rejected": -12.972227096557617, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6523511823865181, | |
| "eval_logits/chosen": -2.528658628463745, | |
| "eval_logits/rejected": -2.896522045135498, | |
| "eval_logps/chosen": -5.151096820831299, | |
| "eval_logps/rejected": -8.659164428710938, | |
| "eval_loss": 0.27813127636909485, | |
| "eval_rewards/accuracies": 0.9017857313156128, | |
| "eval_rewards/chosen": -7.7266459465026855, | |
| "eval_rewards/margins": 5.262101650238037, | |
| "eval_rewards/rejected": -12.988746643066406, | |
| "eval_runtime": 30.2603, | |
| "eval_samples_per_second": 29.478, | |
| "eval_steps_per_second": 3.701, | |
| "step": 900 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.342232258554626e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |