{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0, "eval_steps": 100, "global_step": 478, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 1.0416666666666666e-08, "logits/chosen": -2.5017967224121094, "logits/rejected": -2.3871021270751953, "logps/chosen": -332.3011474609375, "logps/rejected": -277.1512756347656, "loss": 0.6931, "rewards/accuracies": 0.0, "rewards/chosen": 0.0, "rewards/margins": 0.0, "rewards/rejected": 0.0, "step": 1 }, { "epoch": 0.02, "learning_rate": 1.0416666666666667e-07, "logits/chosen": -2.540015459060669, "logits/rejected": -2.500929832458496, "logps/chosen": -247.11024475097656, "logps/rejected": -213.2850341796875, "loss": 0.6932, "rewards/accuracies": 0.4097222089767456, "rewards/chosen": -0.000517037755344063, "rewards/margins": -0.00017402732919435948, "rewards/rejected": -0.0003430104407016188, "step": 10 }, { "epoch": 0.04, "learning_rate": 2.0833333333333333e-07, "logits/chosen": -2.5355143547058105, "logits/rejected": -2.484562873840332, "logps/chosen": -272.35711669921875, "logps/rejected": -249.6931915283203, "loss": 0.6924, "rewards/accuracies": 0.581250011920929, "rewards/chosen": 0.0008343200897797942, "rewards/margins": 0.001956597436219454, "rewards/rejected": -0.001122277113609016, "step": 20 }, { "epoch": 0.06, "learning_rate": 3.1249999999999997e-07, "logits/chosen": -2.6150834560394287, "logits/rejected": -2.542232036590576, "logps/chosen": -283.331298828125, "logps/rejected": -276.0316467285156, "loss": 0.6885, "rewards/accuracies": 0.612500011920929, "rewards/chosen": 0.008794735185801983, "rewards/margins": 0.006988237611949444, "rewards/rejected": 0.0018064973410218954, "step": 30 }, { "epoch": 0.08, "learning_rate": 4.1666666666666667e-07, "logits/chosen": -2.519700527191162, "logits/rejected": -2.4544477462768555, "logps/chosen": -288.609130859375, "logps/rejected": -265.47576904296875, "loss": 0.6756, "rewards/accuracies": 0.668749988079071, "rewards/chosen": 0.04671400785446167, "rewards/margins": 0.03756200894713402, "rewards/rejected": 0.009151997044682503, "step": 40 }, { "epoch": 0.1, "learning_rate": 4.999733114418725e-07, "logits/chosen": -2.4968419075012207, "logits/rejected": -2.4539003372192383, "logps/chosen": -312.46380615234375, "logps/rejected": -309.41632080078125, "loss": 0.6601, "rewards/accuracies": 0.7250000238418579, "rewards/chosen": -0.006021331064403057, "rewards/margins": 0.07272790372371674, "rewards/rejected": -0.07874923944473267, "step": 50 }, { "epoch": 0.13, "learning_rate": 4.990398100856366e-07, "logits/chosen": -2.461306095123291, "logits/rejected": -2.4000632762908936, "logps/chosen": -267.9833679199219, "logps/rejected": -251.8109130859375, "loss": 0.6381, "rewards/accuracies": 0.668749988079071, "rewards/chosen": -0.10297150909900665, "rewards/margins": 0.12468205392360687, "rewards/rejected": -0.22765357792377472, "step": 60 }, { "epoch": 0.15, "learning_rate": 4.967775735898179e-07, "logits/chosen": -2.427084445953369, "logits/rejected": -2.366863250732422, "logps/chosen": -298.81085205078125, "logps/rejected": -287.7050476074219, "loss": 0.6177, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -0.1814979612827301, "rewards/margins": 0.2026226818561554, "rewards/rejected": -0.3841206133365631, "step": 70 }, { "epoch": 0.17, "learning_rate": 4.931986719649298e-07, "logits/chosen": -1.6477082967758179, "logits/rejected": -1.5141746997833252, "logps/chosen": -328.52178955078125, "logps/rejected": -357.9725646972656, "loss": 0.609, "rewards/accuracies": 0.699999988079071, "rewards/chosen": -0.5934440493583679, "rewards/margins": 0.24087591469287872, "rewards/rejected": -0.8343199491500854, "step": 80 }, { "epoch": 0.19, "learning_rate": 4.883222001996351e-07, "logits/chosen": -1.456400990486145, "logits/rejected": -1.1816024780273438, "logps/chosen": -324.58734130859375, "logps/rejected": -347.7338562011719, "loss": 0.5717, "rewards/accuracies": 0.7124999761581421, "rewards/chosen": -0.5771389603614807, "rewards/margins": 0.38918066024780273, "rewards/rejected": -0.966319739818573, "step": 90 }, { "epoch": 0.21, "learning_rate": 4.821741763807186e-07, "logits/chosen": -1.2872960567474365, "logits/rejected": -0.9513088464736938, "logps/chosen": -339.1795959472656, "logps/rejected": -366.32769775390625, "loss": 0.5545, "rewards/accuracies": 0.699999988079071, "rewards/chosen": -0.5583249926567078, "rewards/margins": 0.47431764006614685, "rewards/rejected": -1.0326426029205322, "step": 100 }, { "epoch": 0.21, "eval_logits/chosen": -1.3212240934371948, "eval_logits/rejected": -1.0286760330200195, "eval_logps/chosen": -312.0798645019531, "eval_logps/rejected": -374.31585693359375, "eval_loss": 0.5658453106880188, "eval_rewards/accuracies": 0.71875, "eval_rewards/chosen": -0.49532508850097656, "eval_rewards/margins": 0.6263692378997803, "eval_rewards/rejected": -1.1216944456100464, "eval_runtime": 88.7193, "eval_samples_per_second": 22.543, "eval_steps_per_second": 0.361, "step": 100 }, { "epoch": 0.23, "learning_rate": 4.747874028753375e-07, "logits/chosen": -1.1495229005813599, "logits/rejected": -0.6862327456474304, "logps/chosen": -370.4473571777344, "logps/rejected": -381.8430480957031, "loss": 0.5534, "rewards/accuracies": 0.75, "rewards/chosen": -0.503128170967102, "rewards/margins": 0.6164419054985046, "rewards/rejected": -1.1195701360702515, "step": 110 }, { "epoch": 0.25, "learning_rate": 4.662012913161997e-07, "logits/chosen": -0.21524420380592346, "logits/rejected": 0.5560011863708496, "logps/chosen": -363.287353515625, "logps/rejected": -385.227783203125, "loss": 0.5308, "rewards/accuracies": 0.71875, "rewards/chosen": -0.8372129201889038, "rewards/margins": 0.710771918296814, "rewards/rejected": -1.5479847192764282, "step": 120 }, { "epoch": 0.27, "learning_rate": 4.5646165232345103e-07, "logits/chosen": -0.9693559408187866, "logits/rejected": -0.585214376449585, "logps/chosen": -310.8514404296875, "logps/rejected": -407.11956787109375, "loss": 0.5387, "rewards/accuracies": 0.7437499761581421, "rewards/chosen": -0.4881950318813324, "rewards/margins": 0.6856316328048706, "rewards/rejected": -1.1738265752792358, "step": 130 }, { "epoch": 0.29, "learning_rate": 4.456204510851956e-07, "logits/chosen": -1.0192829370498657, "logits/rejected": -0.2164110690355301, "logps/chosen": -359.496826171875, "logps/rejected": -383.21826171875, "loss": 0.5381, "rewards/accuracies": 0.7562500238418579, "rewards/chosen": -0.6586028337478638, "rewards/margins": 0.6349435448646545, "rewards/rejected": -1.2935463190078735, "step": 140 }, { "epoch": 0.31, "learning_rate": 4.337355301007335e-07, "logits/chosen": -0.4600129723548889, "logits/rejected": 0.2966030240058899, "logps/chosen": -379.4418640136719, "logps/rejected": -396.3006286621094, "loss": 0.5288, "rewards/accuracies": 0.7124999761581421, "rewards/chosen": -0.9255654215812683, "rewards/margins": 0.6836373805999756, "rewards/rejected": -1.6092027425765991, "step": 150 }, { "epoch": 0.33, "learning_rate": 4.2087030056579986e-07, "logits/chosen": -0.17871518433094025, "logits/rejected": 0.4601938724517822, "logps/chosen": -382.32659912109375, "logps/rejected": -447.52716064453125, "loss": 0.5063, "rewards/accuracies": 0.7875000238418579, "rewards/chosen": -0.9520798921585083, "rewards/margins": 0.810064435005188, "rewards/rejected": -1.7621443271636963, "step": 160 }, { "epoch": 0.36, "learning_rate": 4.070934040463998e-07, "logits/chosen": -0.7312272191047668, "logits/rejected": -0.21112406253814697, "logps/chosen": -320.51849365234375, "logps/rejected": -360.88427734375, "loss": 0.5618, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -0.7297524213790894, "rewards/margins": 0.565701425075531, "rewards/rejected": -1.2954537868499756, "step": 170 }, { "epoch": 0.38, "learning_rate": 3.9247834624635404e-07, "logits/chosen": -0.5727821588516235, "logits/rejected": -0.19889096915721893, "logps/chosen": -369.8377990722656, "logps/rejected": -436.2527770996094, "loss": 0.5238, "rewards/accuracies": 0.71875, "rewards/chosen": -0.7431085705757141, "rewards/margins": 0.8097877502441406, "rewards/rejected": -1.55289626121521, "step": 180 }, { "epoch": 0.4, "learning_rate": 3.7710310482256523e-07, "logits/chosen": -0.8184123039245605, "logits/rejected": -0.09880775213241577, "logps/chosen": -360.60174560546875, "logps/rejected": -434.35540771484375, "loss": 0.5039, "rewards/accuracies": 0.762499988079071, "rewards/chosen": -0.7761167287826538, "rewards/margins": 0.8615070581436157, "rewards/rejected": -1.6376237869262695, "step": 190 }, { "epoch": 0.42, "learning_rate": 3.610497133404795e-07, "logits/chosen": -0.597638726234436, "logits/rejected": 0.14091506600379944, "logps/chosen": -373.0635681152344, "logps/rejected": -402.6120910644531, "loss": 0.5026, "rewards/accuracies": 0.65625, "rewards/chosen": -0.908000648021698, "rewards/margins": 0.6652692556381226, "rewards/rejected": -1.5732697248458862, "step": 200 }, { "epoch": 0.42, "eval_logits/chosen": -0.17727772891521454, "eval_logits/rejected": 0.5189895629882812, "eval_logps/chosen": -352.49847412109375, "eval_logps/rejected": -439.3264465332031, "eval_loss": 0.5201631188392639, "eval_rewards/accuracies": 0.74609375, "eval_rewards/chosen": -0.899510383605957, "eval_rewards/margins": 0.8722902536392212, "eval_rewards/rejected": -1.7718006372451782, "eval_runtime": 88.5484, "eval_samples_per_second": 22.587, "eval_steps_per_second": 0.361, "step": 200 }, { "epoch": 0.44, "learning_rate": 3.4440382358952115e-07, "logits/chosen": -0.04346243292093277, "logits/rejected": 0.8316682577133179, "logps/chosen": -377.2864990234375, "logps/rejected": -403.12933349609375, "loss": 0.5357, "rewards/accuracies": 0.6937500238418579, "rewards/chosen": -1.005774736404419, "rewards/margins": 0.6370893716812134, "rewards/rejected": -1.6428638696670532, "step": 210 }, { "epoch": 0.46, "learning_rate": 3.272542485937368e-07, "logits/chosen": 0.27445563673973083, "logits/rejected": 0.8739471435546875, "logps/chosen": -359.21343994140625, "logps/rejected": -410.8561096191406, "loss": 0.5269, "rewards/accuracies": 0.793749988079071, "rewards/chosen": -0.8274116516113281, "rewards/margins": 0.8387683033943176, "rewards/rejected": -1.6661800146102905, "step": 220 }, { "epoch": 0.48, "learning_rate": 3.096924887558854e-07, "logits/chosen": 0.12390404939651489, "logits/rejected": 0.7874934673309326, "logps/chosen": -343.6993713378906, "logps/rejected": -443.02716064453125, "loss": 0.5192, "rewards/accuracies": 0.7875000238418579, "rewards/chosen": -0.7115140557289124, "rewards/margins": 0.9087392091751099, "rewards/rejected": -1.6202532052993774, "step": 230 }, { "epoch": 0.5, "learning_rate": 2.9181224366319943e-07, "logits/chosen": 0.47393637895584106, "logits/rejected": 1.2820873260498047, "logps/chosen": -341.52032470703125, "logps/rejected": -402.3468322753906, "loss": 0.5251, "rewards/accuracies": 0.699999988079071, "rewards/chosen": -0.9017995595932007, "rewards/margins": 0.7208075523376465, "rewards/rejected": -1.6226072311401367, "step": 240 }, { "epoch": 0.52, "learning_rate": 2.7370891215954565e-07, "logits/chosen": 0.6556827425956726, "logits/rejected": 1.5371648073196411, "logps/chosen": -363.3061218261719, "logps/rejected": -423.22216796875, "loss": 0.5168, "rewards/accuracies": 0.6937500238418579, "rewards/chosen": -0.9660114049911499, "rewards/margins": 0.6768711805343628, "rewards/rejected": -1.6428825855255127, "step": 250 }, { "epoch": 0.54, "learning_rate": 2.55479083351317e-07, "logits/chosen": 1.074449062347412, "logits/rejected": 2.2258691787719727, "logps/chosen": -393.25042724609375, "logps/rejected": -435.5625915527344, "loss": 0.5007, "rewards/accuracies": 0.793749988079071, "rewards/chosen": -1.1183383464813232, "rewards/margins": 0.817892849445343, "rewards/rejected": -1.936231255531311, "step": 260 }, { "epoch": 0.56, "learning_rate": 2.3722002126275822e-07, "logits/chosen": 1.035612940788269, "logits/rejected": 1.9490232467651367, "logps/chosen": -391.06488037109375, "logps/rejected": -470.64599609375, "loss": 0.5064, "rewards/accuracies": 0.7749999761581421, "rewards/chosen": -1.1720463037490845, "rewards/margins": 0.9149090051651001, "rewards/rejected": -2.0869553089141846, "step": 270 }, { "epoch": 0.59, "learning_rate": 2.19029145890313e-07, "logits/chosen": 0.5210274457931519, "logits/rejected": 1.496964693069458, "logps/chosen": -369.4839782714844, "logps/rejected": -415.17974853515625, "loss": 0.5189, "rewards/accuracies": 0.762499988079071, "rewards/chosen": -0.9482043981552124, "rewards/margins": 0.8109124898910522, "rewards/rejected": -1.759116768836975, "step": 280 }, { "epoch": 0.61, "learning_rate": 2.0100351342479216e-07, "logits/chosen": 0.4483126103878021, "logits/rejected": 1.3368585109710693, "logps/chosen": -373.81475830078125, "logps/rejected": -435.36151123046875, "loss": 0.519, "rewards/accuracies": 0.7437499761581421, "rewards/chosen": -0.8730584979057312, "rewards/margins": 0.7535277009010315, "rewards/rejected": -1.6265861988067627, "step": 290 }, { "epoch": 0.63, "learning_rate": 1.8323929841460178e-07, "logits/chosen": 0.27887973189353943, "logits/rejected": 1.0783421993255615, "logps/chosen": -381.5480651855469, "logps/rejected": -453.80242919921875, "loss": 0.5106, "rewards/accuracies": 0.7250000238418579, "rewards/chosen": -0.8568315505981445, "rewards/margins": 0.6526933312416077, "rewards/rejected": -1.5095248222351074, "step": 300 }, { "epoch": 0.63, "eval_logits/chosen": 0.08615332096815109, "eval_logits/rejected": 0.9099248647689819, "eval_logps/chosen": -342.00433349609375, "eval_logps/rejected": -424.99755859375, "eval_loss": 0.5103623270988464, "eval_rewards/accuracies": 0.765625, "eval_rewards/chosen": -0.7945692539215088, "eval_rewards/margins": 0.8339425921440125, "eval_rewards/rejected": -1.628511905670166, "eval_runtime": 88.6653, "eval_samples_per_second": 22.557, "eval_steps_per_second": 0.361, "step": 300 }, { "epoch": 0.65, "learning_rate": 1.6583128063291573e-07, "logits/chosen": 0.13822266459465027, "logits/rejected": 1.3017876148223877, "logps/chosen": -365.63958740234375, "logps/rejected": -384.9111328125, "loss": 0.4988, "rewards/accuracies": 0.75, "rewards/chosen": -0.9199568629264832, "rewards/margins": 0.7403375506401062, "rewards/rejected": -1.6602942943572998, "step": 310 }, { "epoch": 0.67, "learning_rate": 1.488723393865766e-07, "logits/chosen": 0.672313392162323, "logits/rejected": 1.8797632455825806, "logps/chosen": -326.7810974121094, "logps/rejected": -421.84765625, "loss": 0.5073, "rewards/accuracies": 0.7749999761581421, "rewards/chosen": -0.7888548374176025, "rewards/margins": 1.0594618320465088, "rewards/rejected": -1.8483164310455322, "step": 320 }, { "epoch": 0.69, "learning_rate": 1.3245295796480788e-07, "logits/chosen": 0.46894121170043945, "logits/rejected": 1.3748019933700562, "logps/chosen": -373.7689208984375, "logps/rejected": -419.54364013671875, "loss": 0.5143, "rewards/accuracies": 0.7562500238418579, "rewards/chosen": -0.8813614845275879, "rewards/margins": 0.7780431509017944, "rewards/rejected": -1.6594045162200928, "step": 330 }, { "epoch": 0.71, "learning_rate": 1.1666074087171627e-07, "logits/chosen": 0.8704012036323547, "logits/rejected": 1.6405874490737915, "logps/chosen": -351.2308654785156, "logps/rejected": -407.41064453125, "loss": 0.5135, "rewards/accuracies": 0.7250000238418579, "rewards/chosen": -0.8790988922119141, "rewards/margins": 0.7077994346618652, "rewards/rejected": -1.5868984460830688, "step": 340 }, { "epoch": 0.73, "learning_rate": 1.0157994641835734e-07, "logits/chosen": 0.8922025561332703, "logits/rejected": 2.06630277633667, "logps/chosen": -348.2925720214844, "logps/rejected": -419.91363525390625, "loss": 0.4771, "rewards/accuracies": 0.8125, "rewards/chosen": -0.9759310483932495, "rewards/margins": 0.898769736289978, "rewards/rejected": -1.8747007846832275, "step": 350 }, { "epoch": 0.75, "learning_rate": 8.729103716819111e-08, "logits/chosen": 0.996601402759552, "logits/rejected": 2.303062915802002, "logps/chosen": -422.4617614746094, "logps/rejected": -457.77691650390625, "loss": 0.4833, "rewards/accuracies": 0.7749999761581421, "rewards/chosen": -1.044933557510376, "rewards/margins": 0.9208766222000122, "rewards/rejected": -1.9658104181289673, "step": 360 }, { "epoch": 0.77, "learning_rate": 7.387025063449081e-08, "logits/chosen": 1.0122365951538086, "logits/rejected": 2.279644727706909, "logps/chosen": -411.8594665527344, "logps/rejected": -457.96807861328125, "loss": 0.4969, "rewards/accuracies": 0.8062499761581421, "rewards/chosen": -1.0128066539764404, "rewards/margins": 1.0020781755447388, "rewards/rejected": -2.0148847103118896, "step": 370 }, { "epoch": 0.79, "learning_rate": 6.138919252022435e-08, "logits/chosen": 0.8450608253479004, "logits/rejected": 1.820166826248169, "logps/chosen": -383.67559814453125, "logps/rejected": -475.4779357910156, "loss": 0.4768, "rewards/accuracies": 0.768750011920929, "rewards/chosen": -1.0418307781219482, "rewards/margins": 0.9196658134460449, "rewards/rejected": -1.9614967107772827, "step": 380 }, { "epoch": 0.82, "learning_rate": 4.991445467064689e-08, "logits/chosen": 1.0426132678985596, "logits/rejected": 1.9015953540802002, "logps/chosen": -384.1719970703125, "logps/rejected": -459.9111328125, "loss": 0.5073, "rewards/accuracies": 0.6937500238418579, "rewards/chosen": -1.1213358640670776, "rewards/margins": 0.7478152513504028, "rewards/rejected": -1.8691511154174805, "step": 390 }, { "epoch": 0.84, "learning_rate": 3.9507259776993954e-08, "logits/chosen": 0.8149229288101196, "logits/rejected": 1.6555734872817993, "logps/chosen": -430.68231201171875, "logps/rejected": -496.66900634765625, "loss": 0.4859, "rewards/accuracies": 0.7749999761581421, "rewards/chosen": -1.098642110824585, "rewards/margins": 0.9415690302848816, "rewards/rejected": -2.0402112007141113, "step": 400 }, { "epoch": 0.84, "eval_logits/chosen": 0.7817753553390503, "eval_logits/rejected": 1.7438044548034668, "eval_logps/chosen": -360.3138732910156, "eval_logps/rejected": -457.9451599121094, "eval_loss": 0.5031262040138245, "eval_rewards/accuracies": 0.7578125, "eval_rewards/chosen": -0.9776647090911865, "eval_rewards/margins": 0.9803228974342346, "eval_rewards/rejected": -1.9579875469207764, "eval_runtime": 88.4827, "eval_samples_per_second": 22.603, "eval_steps_per_second": 0.362, "step": 400 }, { "epoch": 0.86, "learning_rate": 3.022313472693447e-08, "logits/chosen": 1.0873724222183228, "logits/rejected": 2.1809914112091064, "logps/chosen": -358.1825256347656, "logps/rejected": -440.631591796875, "loss": 0.5114, "rewards/accuracies": 0.762499988079071, "rewards/chosen": -0.9868854284286499, "rewards/margins": 0.9311205744743347, "rewards/rejected": -1.918006181716919, "step": 410 }, { "epoch": 0.88, "learning_rate": 2.2111614344599684e-08, "logits/chosen": 0.8678689002990723, "logits/rejected": 1.581181526184082, "logps/chosen": -370.5611877441406, "logps/rejected": -477.7791442871094, "loss": 0.5041, "rewards/accuracies": 0.75, "rewards/chosen": -0.9568517804145813, "rewards/margins": 0.967370331287384, "rewards/rejected": -1.9242219924926758, "step": 420 }, { "epoch": 0.9, "learning_rate": 1.521597710086439e-08, "logits/chosen": 0.4731677174568176, "logits/rejected": 1.4058442115783691, "logps/chosen": -396.15325927734375, "logps/rejected": -442.45404052734375, "loss": 0.5063, "rewards/accuracies": 0.762499988079071, "rewards/chosen": -0.994260311126709, "rewards/margins": 0.8869916796684265, "rewards/rejected": -1.8812516927719116, "step": 430 }, { "epoch": 0.92, "learning_rate": 9.57301420397924e-09, "logits/chosen": 0.8329499959945679, "logits/rejected": 1.7063286304473877, "logps/chosen": -360.3570861816406, "logps/rejected": -442.45574951171875, "loss": 0.4884, "rewards/accuracies": 0.762499988079071, "rewards/chosen": -1.0168806314468384, "rewards/margins": 0.8515257835388184, "rewards/rejected": -1.8684062957763672, "step": 440 }, { "epoch": 0.94, "learning_rate": 5.212833302556258e-09, "logits/chosen": 0.8873745799064636, "logits/rejected": 1.8809913396835327, "logps/chosen": -344.71197509765625, "logps/rejected": -413.39471435546875, "loss": 0.5068, "rewards/accuracies": 0.7437499761581421, "rewards/chosen": -0.99580317735672, "rewards/margins": 0.8256435394287109, "rewards/rejected": -1.8214466571807861, "step": 450 }, { "epoch": 0.96, "learning_rate": 2.158697848236607e-09, "logits/chosen": 1.0402039289474487, "logits/rejected": 2.0626494884490967, "logps/chosen": -350.25115966796875, "logps/rejected": -416.1396484375, "loss": 0.4859, "rewards/accuracies": 0.737500011920929, "rewards/chosen": -1.070770502090454, "rewards/margins": 0.807015597820282, "rewards/rejected": -1.8777860403060913, "step": 460 }, { "epoch": 0.98, "learning_rate": 4.269029751107489e-10, "logits/chosen": 0.4914283752441406, "logits/rejected": 1.8748416900634766, "logps/chosen": -398.6361083984375, "logps/rejected": -465.1482849121094, "loss": 0.4872, "rewards/accuracies": 0.800000011920929, "rewards/chosen": -1.004058599472046, "rewards/margins": 1.027204155921936, "rewards/rejected": -2.0312628746032715, "step": 470 }, { "epoch": 1.0, "step": 478, "total_flos": 0.0, "train_loss": 0.07998354963677698, "train_runtime": 779.5821, "train_samples_per_second": 78.42, "train_steps_per_second": 0.613 } ], "logging_steps": 10, "max_steps": 478, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 100, "total_flos": 0.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }