| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9981298423724285, | |
| "eval_steps": 400, | |
| "global_step": 467, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021373230029388193, | |
| "grad_norm": 14.616270006854805, | |
| "learning_rate": 2.127659574468085e-08, | |
| "logits/chosen": -1.1381689310073853, | |
| "logits/rejected": -0.9913416504859924, | |
| "logps/chosen": -0.2839311361312866, | |
| "logps/rejected": -0.29555341601371765, | |
| "loss": 1.608, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -0.7098277807235718, | |
| "rewards/margins": 0.029055725783109665, | |
| "rewards/rejected": -0.7388835549354553, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010686615014694095, | |
| "grad_norm": 15.798664787929345, | |
| "learning_rate": 1.0638297872340425e-07, | |
| "logits/chosen": -0.989432692527771, | |
| "logits/rejected": -0.9188639521598816, | |
| "logps/chosen": -0.26978519558906555, | |
| "logps/rejected": -0.26858454942703247, | |
| "loss": 1.6205, | |
| "rewards/accuracies": 0.546875, | |
| "rewards/chosen": -0.6744629740715027, | |
| "rewards/margins": -0.0030015837401151657, | |
| "rewards/rejected": -0.6714614033699036, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02137323002938819, | |
| "grad_norm": 12.624563810334125, | |
| "learning_rate": 2.127659574468085e-07, | |
| "logits/chosen": -0.9885454177856445, | |
| "logits/rejected": -0.9444629549980164, | |
| "logps/chosen": -0.27237212657928467, | |
| "logps/rejected": -0.2738865911960602, | |
| "loss": 1.6069, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.6809303164482117, | |
| "rewards/margins": 0.0037860602606087923, | |
| "rewards/rejected": -0.6847164630889893, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03205984504408229, | |
| "grad_norm": 8.372527145409705, | |
| "learning_rate": 3.1914893617021275e-07, | |
| "logits/chosen": -0.9621208906173706, | |
| "logits/rejected": -0.9169891476631165, | |
| "logps/chosen": -0.2949882745742798, | |
| "logps/rejected": -0.2834514081478119, | |
| "loss": 1.6049, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7374706864356995, | |
| "rewards/margins": -0.028842147439718246, | |
| "rewards/rejected": -0.7086285352706909, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04274646005877638, | |
| "grad_norm": 11.776984926293357, | |
| "learning_rate": 4.25531914893617e-07, | |
| "logits/chosen": -0.9825040102005005, | |
| "logits/rejected": -0.983070969581604, | |
| "logps/chosen": -0.26186102628707886, | |
| "logps/rejected": -0.26891934871673584, | |
| "loss": 1.6132, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.65465247631073, | |
| "rewards/margins": 0.01764589548110962, | |
| "rewards/rejected": -0.6722984313964844, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.053433075073470476, | |
| "grad_norm": 11.594406861763694, | |
| "learning_rate": 5.319148936170212e-07, | |
| "logits/chosen": -1.0304547548294067, | |
| "logits/rejected": -1.0080630779266357, | |
| "logps/chosen": -0.28237098455429077, | |
| "logps/rejected": -0.28737810254096985, | |
| "loss": 1.6046, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.7059274911880493, | |
| "rewards/margins": 0.012517772614955902, | |
| "rewards/rejected": -0.7184451818466187, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06411969008816458, | |
| "grad_norm": 13.181478771123727, | |
| "learning_rate": 6.382978723404255e-07, | |
| "logits/chosen": -1.0607410669326782, | |
| "logits/rejected": -0.9752001762390137, | |
| "logps/chosen": -0.2747865319252014, | |
| "logps/rejected": -0.27676859498023987, | |
| "loss": 1.6169, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.6869663596153259, | |
| "rewards/margins": 0.004955160431563854, | |
| "rewards/rejected": -0.6919214725494385, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07480630510285867, | |
| "grad_norm": 20.535555442364025, | |
| "learning_rate": 7.446808510638297e-07, | |
| "logits/chosen": -0.9909561276435852, | |
| "logits/rejected": -0.9133442044258118, | |
| "logps/chosen": -0.2775736451148987, | |
| "logps/rejected": -0.3019401431083679, | |
| "loss": 1.6019, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.6939341425895691, | |
| "rewards/margins": 0.06091625243425369, | |
| "rewards/rejected": -0.754850447177887, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08549292011755276, | |
| "grad_norm": 24.246139127435896, | |
| "learning_rate": 8.51063829787234e-07, | |
| "logits/chosen": -0.9217761158943176, | |
| "logits/rejected": -0.9104591608047485, | |
| "logps/chosen": -0.2786272466182709, | |
| "logps/rejected": -0.29105645418167114, | |
| "loss": 1.617, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.6965680718421936, | |
| "rewards/margins": 0.031073052436113358, | |
| "rewards/rejected": -0.7276411056518555, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09617953513224686, | |
| "grad_norm": 10.340400390143392, | |
| "learning_rate": 9.574468085106384e-07, | |
| "logits/chosen": -0.918626606464386, | |
| "logits/rejected": -0.837913990020752, | |
| "logps/chosen": -0.33493560552597046, | |
| "logps/rejected": -0.33847588300704956, | |
| "loss": 1.6012, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.8373388051986694, | |
| "rewards/margins": 0.00885077752172947, | |
| "rewards/rejected": -0.8461896777153015, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10686615014694095, | |
| "grad_norm": 9.374774032732045, | |
| "learning_rate": 9.998741174712533e-07, | |
| "logits/chosen": -0.9050552248954773, | |
| "logits/rejected": -0.9147823452949524, | |
| "logps/chosen": -0.29951199889183044, | |
| "logps/rejected": -0.33792784810066223, | |
| "loss": 1.5734, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.7487800717353821, | |
| "rewards/margins": 0.09603960067033768, | |
| "rewards/rejected": -0.8448196649551392, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11755276516163506, | |
| "grad_norm": 12.319781154359692, | |
| "learning_rate": 9.991050648838675e-07, | |
| "logits/chosen": -0.917604923248291, | |
| "logits/rejected": -0.859523594379425, | |
| "logps/chosen": -0.2723899185657501, | |
| "logps/rejected": -0.3260456621646881, | |
| "loss": 1.5698, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.6809747815132141, | |
| "rewards/margins": 0.13413934409618378, | |
| "rewards/rejected": -0.8151141405105591, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.12823938017632916, | |
| "grad_norm": 10.399641038869387, | |
| "learning_rate": 9.97637968732563e-07, | |
| "logits/chosen": -0.9515789151191711, | |
| "logits/rejected": -0.9302376508712769, | |
| "logps/chosen": -0.28720152378082275, | |
| "logps/rejected": -0.3024197816848755, | |
| "loss": 1.5846, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.7180038690567017, | |
| "rewards/margins": 0.03804563358426094, | |
| "rewards/rejected": -0.7560494542121887, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13892599519102325, | |
| "grad_norm": 11.022271869662724, | |
| "learning_rate": 9.954748808839674e-07, | |
| "logits/chosen": -0.943594753742218, | |
| "logits/rejected": -0.9900406002998352, | |
| "logps/chosen": -0.28493180871009827, | |
| "logps/rejected": -0.28960293531417847, | |
| "loss": 1.584, | |
| "rewards/accuracies": 0.4000000059604645, | |
| "rewards/chosen": -0.7123295664787292, | |
| "rewards/margins": 0.01167784072458744, | |
| "rewards/rejected": -0.7240074276924133, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.14961261020571734, | |
| "grad_norm": 16.92128796829008, | |
| "learning_rate": 9.926188266120295e-07, | |
| "logits/chosen": -0.9900282025337219, | |
| "logits/rejected": -0.9590619802474976, | |
| "logps/chosen": -0.347956120967865, | |
| "logps/rejected": -0.3921958804130554, | |
| "loss": 1.5789, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.8698902130126953, | |
| "rewards/margins": 0.11059943586587906, | |
| "rewards/rejected": -0.9804896116256714, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.16029922522041143, | |
| "grad_norm": 16.614347339411598, | |
| "learning_rate": 9.890738003669027e-07, | |
| "logits/chosen": -0.9594413042068481, | |
| "logits/rejected": -0.9471073150634766, | |
| "logps/chosen": -0.3611677289009094, | |
| "logps/rejected": -0.4440728724002838, | |
| "loss": 1.5386, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.9029192924499512, | |
| "rewards/margins": 0.20726287364959717, | |
| "rewards/rejected": -1.1101821660995483, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.17098584023510552, | |
| "grad_norm": 11.76758635779136, | |
| "learning_rate": 9.848447601883433e-07, | |
| "logits/chosen": -0.9906526803970337, | |
| "logits/rejected": -0.9735655784606934, | |
| "logps/chosen": -0.32699793577194214, | |
| "logps/rejected": -0.40725231170654297, | |
| "loss": 1.5433, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.8174948692321777, | |
| "rewards/margins": 0.20063595473766327, | |
| "rewards/rejected": -1.018130898475647, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.18167245524979964, | |
| "grad_norm": 10.051730020561614, | |
| "learning_rate": 9.799376207714444e-07, | |
| "logits/chosen": -0.9377920031547546, | |
| "logits/rejected": -0.889153003692627, | |
| "logps/chosen": -0.35584911704063416, | |
| "logps/rejected": -0.403181254863739, | |
| "loss": 1.5549, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.8896228671073914, | |
| "rewards/margins": 0.11833026260137558, | |
| "rewards/rejected": -1.0079530477523804, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.19235907026449373, | |
| "grad_norm": 12.299894230325533, | |
| "learning_rate": 9.743592451943998e-07, | |
| "logits/chosen": -0.9021016955375671, | |
| "logits/rejected": -0.8429332971572876, | |
| "logps/chosen": -0.3575811982154846, | |
| "logps/rejected": -0.5039599537849426, | |
| "loss": 1.5625, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.8939528465270996, | |
| "rewards/margins": 0.36594703793525696, | |
| "rewards/rejected": -1.2598999738693237, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 14.145971876119047, | |
| "learning_rate": 9.681174353198686e-07, | |
| "logits/chosen": -1.002209186553955, | |
| "logits/rejected": -1.0227996110916138, | |
| "logps/chosen": -0.41834840178489685, | |
| "logps/rejected": -0.47105270624160767, | |
| "loss": 1.5299, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.0458710193634033, | |
| "rewards/margins": 0.13176079094409943, | |
| "rewards/rejected": -1.1776319742202759, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.2137323002938819, | |
| "grad_norm": 16.666230206302217, | |
| "learning_rate": 9.612209208833646e-07, | |
| "logits/chosen": -1.0650508403778076, | |
| "logits/rejected": -1.0104029178619385, | |
| "logps/chosen": -0.4353299140930176, | |
| "logps/rejected": -0.4307008385658264, | |
| "loss": 1.5889, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -1.088324785232544, | |
| "rewards/margins": -0.011572673916816711, | |
| "rewards/rejected": -1.076751947402954, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.224418915308576, | |
| "grad_norm": 17.277037383709768, | |
| "learning_rate": 9.536793472839324e-07, | |
| "logits/chosen": -0.9813326597213745, | |
| "logits/rejected": -0.9346880912780762, | |
| "logps/chosen": -0.3515966534614563, | |
| "logps/rejected": -0.46131715178489685, | |
| "loss": 1.5543, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.8789916038513184, | |
| "rewards/margins": 0.2743012309074402, | |
| "rewards/rejected": -1.1532928943634033, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.2351055303232701, | |
| "grad_norm": 12.743189576436166, | |
| "learning_rate": 9.455032620941839e-07, | |
| "logits/chosen": -0.8806193470954895, | |
| "logits/rejected": -0.8356341123580933, | |
| "logps/chosen": -0.39990124106407166, | |
| "logps/rejected": -0.42553144693374634, | |
| "loss": 1.5602, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.9997529983520508, | |
| "rewards/margins": 0.06407558172941208, | |
| "rewards/rejected": -1.063828706741333, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2457921453379642, | |
| "grad_norm": 13.62783373071176, | |
| "learning_rate": 9.367041003085648e-07, | |
| "logits/chosen": -0.9096847772598267, | |
| "logits/rejected": -0.9085425138473511, | |
| "logps/chosen": -0.40596961975097656, | |
| "logps/rejected": -0.4234519898891449, | |
| "loss": 1.5452, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -1.0149240493774414, | |
| "rewards/margins": 0.043705932796001434, | |
| "rewards/rejected": -1.0586299896240234, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.2564787603526583, | |
| "grad_norm": 12.129891786309633, | |
| "learning_rate": 9.272941683504808e-07, | |
| "logits/chosen": -0.9741169810295105, | |
| "logits/rejected": -0.9402868151664734, | |
| "logps/chosen": -0.3675619959831238, | |
| "logps/rejected": -0.4012192189693451, | |
| "loss": 1.5295, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.9189050793647766, | |
| "rewards/margins": 0.08414317667484283, | |
| "rewards/rejected": -1.0030481815338135, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2671653753673524, | |
| "grad_norm": 12.760172929151745, | |
| "learning_rate": 9.172866268606513e-07, | |
| "logits/chosen": -1.014088749885559, | |
| "logits/rejected": -1.007265567779541, | |
| "logps/chosen": -0.47613659501075745, | |
| "logps/rejected": -0.5860797166824341, | |
| "loss": 1.5674, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.1903413534164429, | |
| "rewards/margins": 0.2748578190803528, | |
| "rewards/rejected": -1.4651992321014404, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2778519903820465, | |
| "grad_norm": 14.619403511233878, | |
| "learning_rate": 9.066954722907638e-07, | |
| "logits/chosen": -1.0016984939575195, | |
| "logits/rejected": -1.0274698734283447, | |
| "logps/chosen": -0.4267544746398926, | |
| "logps/rejected": -0.5695111155509949, | |
| "loss": 1.5349, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -1.0668861865997314, | |
| "rewards/margins": 0.35689178109169006, | |
| "rewards/rejected": -1.4237778186798096, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2885386053967406, | |
| "grad_norm": 10.619279201046492, | |
| "learning_rate": 8.955355173281707e-07, | |
| "logits/chosen": -0.9727472066879272, | |
| "logits/rejected": -0.9219390749931335, | |
| "logps/chosen": -0.39321380853652954, | |
| "logps/rejected": -0.4508630633354187, | |
| "loss": 1.4978, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.9830344915390015, | |
| "rewards/margins": 0.1441231220960617, | |
| "rewards/rejected": -1.1271576881408691, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.2992252204114347, | |
| "grad_norm": 15.834369827499174, | |
| "learning_rate": 8.838223701790055e-07, | |
| "logits/chosen": -0.967811107635498, | |
| "logits/rejected": -0.9583989977836609, | |
| "logps/chosen": -0.43714672327041626, | |
| "logps/rejected": -0.5161387920379639, | |
| "loss": 1.5218, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.0928667783737183, | |
| "rewards/margins": 0.19748012721538544, | |
| "rewards/rejected": -1.2903468608856201, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.30991183542612877, | |
| "grad_norm": 12.847642374810391, | |
| "learning_rate": 8.71572412738697e-07, | |
| "logits/chosen": -0.9621770977973938, | |
| "logits/rejected": -0.9567694664001465, | |
| "logps/chosen": -0.43871012330055237, | |
| "logps/rejected": -0.4793704152107239, | |
| "loss": 1.5379, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.0967752933502197, | |
| "rewards/margins": 0.10165063291788101, | |
| "rewards/rejected": -1.1984260082244873, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.32059845044082286, | |
| "grad_norm": 11.935024879482611, | |
| "learning_rate": 8.588027776804058e-07, | |
| "logits/chosen": -0.8888929486274719, | |
| "logits/rejected": -0.8930709958076477, | |
| "logps/chosen": -0.5125688910484314, | |
| "logps/rejected": -0.722069263458252, | |
| "loss": 1.5225, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.2814221382141113, | |
| "rewards/margins": 0.5237509608268738, | |
| "rewards/rejected": -1.8051731586456299, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33128506545551695, | |
| "grad_norm": 16.764961972010557, | |
| "learning_rate": 8.455313244934324e-07, | |
| "logits/chosen": -0.9782350659370422, | |
| "logits/rejected": -0.9396332502365112, | |
| "logps/chosen": -0.5236457586288452, | |
| "logps/rejected": -0.7281317710876465, | |
| "loss": 1.4981, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.309114694595337, | |
| "rewards/margins": 0.5112148523330688, | |
| "rewards/rejected": -1.8203294277191162, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.34197168047021104, | |
| "grad_norm": 11.439957416651609, | |
| "learning_rate": 8.317766145051057e-07, | |
| "logits/chosen": -0.9513300061225891, | |
| "logits/rejected": -0.9507732391357422, | |
| "logps/chosen": -0.4817509651184082, | |
| "logps/rejected": -0.6095650792121887, | |
| "loss": 1.5319, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.2043774127960205, | |
| "rewards/margins": 0.3195350468158722, | |
| "rewards/rejected": -1.5239124298095703, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3526582954849052, | |
| "grad_norm": 15.127711801028411, | |
| "learning_rate": 8.175578849210894e-07, | |
| "logits/chosen": -1.0031640529632568, | |
| "logits/rejected": -0.9406889081001282, | |
| "logps/chosen": -0.5859075784683228, | |
| "logps/rejected": -0.6615421175956726, | |
| "loss": 1.5097, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.4647690057754517, | |
| "rewards/margins": 0.18908634781837463, | |
| "rewards/rejected": -1.6538550853729248, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.36334491049959927, | |
| "grad_norm": 19.457497723556852, | |
| "learning_rate": 8.028950219204099e-07, | |
| "logits/chosen": -0.9729808568954468, | |
| "logits/rejected": -0.9703952670097351, | |
| "logps/chosen": -0.5817127823829651, | |
| "logps/rejected": -0.7540786266326904, | |
| "loss": 1.52, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.4542819261550903, | |
| "rewards/margins": 0.4309147298336029, | |
| "rewards/rejected": -1.8851966857910156, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.37403152551429336, | |
| "grad_norm": 13.673431328128673, | |
| "learning_rate": 7.878085328428368e-07, | |
| "logits/chosen": -0.9736520051956177, | |
| "logits/rejected": -0.9835022687911987, | |
| "logps/chosen": -0.6428834795951843, | |
| "logps/rejected": -0.8455599546432495, | |
| "loss": 1.4591, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.6072088479995728, | |
| "rewards/margins": 0.5066913366317749, | |
| "rewards/rejected": -2.1139001846313477, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.38471814052898745, | |
| "grad_norm": 20.186619677768963, | |
| "learning_rate": 7.723195175075135e-07, | |
| "logits/chosen": -0.9989040493965149, | |
| "logits/rejected": -0.9996326565742493, | |
| "logps/chosen": -0.6329335570335388, | |
| "logps/rejected": -0.7996684312820435, | |
| "loss": 1.5042, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.5823338031768799, | |
| "rewards/margins": 0.41683727502822876, | |
| "rewards/rejected": -1.999171257019043, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.39540475554368154, | |
| "grad_norm": 20.89728549129117, | |
| "learning_rate": 7.564496387029531e-07, | |
| "logits/chosen": -1.0245858430862427, | |
| "logits/rejected": -0.9974483251571655, | |
| "logps/chosen": -0.5426946878433228, | |
| "logps/rejected": -0.765714168548584, | |
| "loss": 1.5357, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.3567368984222412, | |
| "rewards/margins": 0.5575486421585083, | |
| "rewards/rejected": -1.914285659790039, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 14.284349268241145, | |
| "learning_rate": 7.402210918896689e-07, | |
| "logits/chosen": -1.0203882455825806, | |
| "logits/rejected": -1.002201795578003, | |
| "logps/chosen": -0.5420663356781006, | |
| "logps/rejected": -0.8374710083007812, | |
| "loss": 1.4458, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -1.355165719985962, | |
| "rewards/margins": 0.7385115623474121, | |
| "rewards/rejected": -2.093677282333374, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4167779855730697, | |
| "grad_norm": 21.006500802721117, | |
| "learning_rate": 7.236565741578162e-07, | |
| "logits/chosen": -0.9360872507095337, | |
| "logits/rejected": -0.9267364740371704, | |
| "logps/chosen": -0.6582534909248352, | |
| "logps/rejected": -0.9019840359687805, | |
| "loss": 1.496, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.6456336975097656, | |
| "rewards/margins": 0.6093264818191528, | |
| "rewards/rejected": -2.254960298538208, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.4274646005877638, | |
| "grad_norm": 17.34001688115439, | |
| "learning_rate": 7.067792524832603e-07, | |
| "logits/chosen": -0.9733101725578308, | |
| "logits/rejected": -0.9737011194229126, | |
| "logps/chosen": -0.5885142087936401, | |
| "logps/rejected": -0.759198784828186, | |
| "loss": 1.4947, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -1.4712855815887451, | |
| "rewards/margins": 0.42671123147010803, | |
| "rewards/rejected": -1.8979966640472412, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4381512156024579, | |
| "grad_norm": 18.93087179194272, | |
| "learning_rate": 6.896127313264642e-07, | |
| "logits/chosen": -1.0645233392715454, | |
| "logits/rejected": -1.0247756242752075, | |
| "logps/chosen": -0.7058154940605164, | |
| "logps/rejected": -0.8320499658584595, | |
| "loss": 1.481, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.7645390033721924, | |
| "rewards/margins": 0.3155860900878906, | |
| "rewards/rejected": -2.080124855041504, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.448837830617152, | |
| "grad_norm": 20.51304681116505, | |
| "learning_rate": 6.721810196195174e-07, | |
| "logits/chosen": -1.0714137554168701, | |
| "logits/rejected": -0.9989116787910461, | |
| "logps/chosen": -0.6715008020401001, | |
| "logps/rejected": -0.8458214998245239, | |
| "loss": 1.4931, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -1.6787519454956055, | |
| "rewards/margins": 0.43580159544944763, | |
| "rewards/rejected": -2.114553451538086, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.45952444563184613, | |
| "grad_norm": 16.46301637400639, | |
| "learning_rate": 6.545084971874736e-07, | |
| "logits/chosen": -1.0320379734039307, | |
| "logits/rejected": -0.9812155961990356, | |
| "logps/chosen": -0.7699872255325317, | |
| "logps/rejected": -0.9591902494430542, | |
| "loss": 1.4357, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.9249680042266846, | |
| "rewards/margins": 0.47300752997398376, | |
| "rewards/rejected": -2.397975444793701, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.4702110606465402, | |
| "grad_norm": 21.088635750930834, | |
| "learning_rate": 6.3661988065096e-07, | |
| "logits/chosen": -1.0327280759811401, | |
| "logits/rejected": -0.9874359965324402, | |
| "logps/chosen": -0.8896724581718445, | |
| "logps/rejected": -1.0073192119598389, | |
| "loss": 1.4954, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.2241809368133545, | |
| "rewards/margins": 0.2941167950630188, | |
| "rewards/rejected": -2.5182979106903076, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.4808976756612343, | |
| "grad_norm": 19.02142810522417, | |
| "learning_rate": 6.185401888577487e-07, | |
| "logits/chosen": -0.9864276051521301, | |
| "logits/rejected": -0.9660781621932983, | |
| "logps/chosen": -0.772359311580658, | |
| "logps/rejected": -0.794577956199646, | |
| "loss": 1.557, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.9308984279632568, | |
| "rewards/margins": 0.0555465929210186, | |
| "rewards/rejected": -1.9864448308944702, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4915842906759284, | |
| "grad_norm": 17.867436451445265, | |
| "learning_rate": 6.002947078916364e-07, | |
| "logits/chosen": -1.0150645971298218, | |
| "logits/rejected": -0.9232436418533325, | |
| "logps/chosen": -0.8819006085395813, | |
| "logps/rejected": -1.0791282653808594, | |
| "loss": 1.4692, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.2047512531280518, | |
| "rewards/margins": 0.49306946992874146, | |
| "rewards/rejected": -2.6978209018707275, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5022709056906225, | |
| "grad_norm": 19.261147315812543, | |
| "learning_rate": 5.819089557075688e-07, | |
| "logits/chosen": -0.996496319770813, | |
| "logits/rejected": -0.9518272280693054, | |
| "logps/chosen": -0.8440850377082825, | |
| "logps/rejected": -0.9660031199455261, | |
| "loss": 1.4942, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.110212564468384, | |
| "rewards/margins": 0.30479517579078674, | |
| "rewards/rejected": -2.4150078296661377, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.5129575207053166, | |
| "grad_norm": 18.32103206829717, | |
| "learning_rate": 5.634086464424742e-07, | |
| "logits/chosen": -1.0355093479156494, | |
| "logits/rejected": -1.006392002105713, | |
| "logps/chosen": -0.7564712762832642, | |
| "logps/rejected": -0.9403325319290161, | |
| "loss": 1.4046, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.8911781311035156, | |
| "rewards/margins": 0.4596532881259918, | |
| "rewards/rejected": -2.3508315086364746, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5236441357200107, | |
| "grad_norm": 13.624682866914808, | |
| "learning_rate": 5.448196544517167e-07, | |
| "logits/chosen": -1.0392498970031738, | |
| "logits/rejected": -1.0080206394195557, | |
| "logps/chosen": -0.7070968151092529, | |
| "logps/rejected": -0.8767108917236328, | |
| "loss": 1.4407, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -1.7677417993545532, | |
| "rewards/margins": 0.4240352213382721, | |
| "rewards/rejected": -2.191777229309082, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.5343307507347048, | |
| "grad_norm": 22.22843267157582, | |
| "learning_rate": 5.26167978121472e-07, | |
| "logits/chosen": -1.048105001449585, | |
| "logits/rejected": -1.028188943862915, | |
| "logps/chosen": -0.8559715151786804, | |
| "logps/rejected": -1.2052185535430908, | |
| "loss": 1.467, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.1399290561676025, | |
| "rewards/margins": 0.8731171488761902, | |
| "rewards/rejected": -3.0130460262298584, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5450173657493989, | |
| "grad_norm": 21.7502246417763, | |
| "learning_rate": 5.074797035076318e-07, | |
| "logits/chosen": -1.0093282461166382, | |
| "logits/rejected": -1.0085898637771606, | |
| "logps/chosen": -0.8086752891540527, | |
| "logps/rejected": -0.9931262731552124, | |
| "loss": 1.4101, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.021688222885132, | |
| "rewards/margins": 0.4611276090145111, | |
| "rewards/rejected": -2.4828155040740967, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.555703980764093, | |
| "grad_norm": 17.422008542886676, | |
| "learning_rate": 4.887809678520975e-07, | |
| "logits/chosen": -0.9935086369514465, | |
| "logits/rejected": -0.9804226160049438, | |
| "logps/chosen": -0.7468116879463196, | |
| "logps/rejected": -1.0926599502563477, | |
| "loss": 1.4384, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -1.8670291900634766, | |
| "rewards/margins": 0.8646209836006165, | |
| "rewards/rejected": -2.731649875640869, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.566390595778787, | |
| "grad_norm": 21.42359137838811, | |
| "learning_rate": 4.700979230274829e-07, | |
| "logits/chosen": -1.0343437194824219, | |
| "logits/rejected": -1.0285662412643433, | |
| "logps/chosen": -0.8779309391975403, | |
| "logps/rejected": -0.9997223019599915, | |
| "loss": 1.4835, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.19482684135437, | |
| "rewards/margins": 0.3044784963130951, | |
| "rewards/rejected": -2.4993057250976562, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.5770772107934812, | |
| "grad_norm": 27.361650975575174, | |
| "learning_rate": 4.514566989613559e-07, | |
| "logits/chosen": -1.022482991218567, | |
| "logits/rejected": -1.030027151107788, | |
| "logps/chosen": -0.8233085870742798, | |
| "logps/rejected": -1.0775353908538818, | |
| "loss": 1.4746, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.058271646499634, | |
| "rewards/margins": 0.6355669498443604, | |
| "rewards/rejected": -2.693838596343994, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5877638258081752, | |
| "grad_norm": 19.306011967200188, | |
| "learning_rate": 4.328833670911724e-07, | |
| "logits/chosen": -0.8774306178092957, | |
| "logits/rejected": -0.841380774974823, | |
| "logps/chosen": -0.977648138999939, | |
| "logps/rejected": -1.1523730754852295, | |
| "loss": 1.4355, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.444120168685913, | |
| "rewards/margins": 0.43681272864341736, | |
| "rewards/rejected": -2.8809328079223633, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5984504408228694, | |
| "grad_norm": 18.830991297114817, | |
| "learning_rate": 4.144039039010124e-07, | |
| "logits/chosen": -1.0766229629516602, | |
| "logits/rejected": -1.034156084060669, | |
| "logps/chosen": -0.9281458854675293, | |
| "logps/rejected": -1.0838878154754639, | |
| "loss": 1.4465, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.3203647136688232, | |
| "rewards/margins": 0.38935500383377075, | |
| "rewards/rejected": -2.709719657897949, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 20.938307778539542, | |
| "learning_rate": 3.960441545911204e-07, | |
| "logits/chosen": -0.9466499090194702, | |
| "logits/rejected": -0.9020398855209351, | |
| "logps/chosen": -1.1035289764404297, | |
| "logps/rejected": -1.1428934335708618, | |
| "loss": 1.4739, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.758822441101074, | |
| "rewards/margins": 0.09841099381446838, | |
| "rewards/rejected": -2.8572330474853516, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.6198236708522575, | |
| "grad_norm": 23.245653263780497, | |
| "learning_rate": 3.778297969310529e-07, | |
| "logits/chosen": -0.9969805479049683, | |
| "logits/rejected": -0.9768760800361633, | |
| "logps/chosen": -0.9135320782661438, | |
| "logps/rejected": -1.1653510332107544, | |
| "loss": 1.3943, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.283830165863037, | |
| "rewards/margins": 0.6295474767684937, | |
| "rewards/rejected": -2.9133777618408203, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6305102858669517, | |
| "grad_norm": 18.66753481518412, | |
| "learning_rate": 3.5978630534699865e-07, | |
| "logits/chosen": -1.0683342218399048, | |
| "logits/rejected": -1.0506360530853271, | |
| "logps/chosen": -1.0534954071044922, | |
| "logps/rejected": -1.2509009838104248, | |
| "loss": 1.4446, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.6337387561798096, | |
| "rewards/margins": 0.4935137629508972, | |
| "rewards/rejected": -3.1272525787353516, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.6411969008816457, | |
| "grad_norm": 21.410873330760186, | |
| "learning_rate": 3.4193891529348795e-07, | |
| "logits/chosen": -0.943785548210144, | |
| "logits/rejected": -0.8960458040237427, | |
| "logps/chosen": -0.9090589284896851, | |
| "logps/rejected": -1.2448195219039917, | |
| "loss": 1.4048, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.2726473808288574, | |
| "rewards/margins": 0.8394016027450562, | |
| "rewards/rejected": -3.112048625946045, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6518835158963399, | |
| "grad_norm": 16.99902973812307, | |
| "learning_rate": 3.243125879593286e-07, | |
| "logits/chosen": -1.001460075378418, | |
| "logits/rejected": -0.9785451889038086, | |
| "logps/chosen": -1.0239307880401611, | |
| "logps/rejected": -1.286873698234558, | |
| "loss": 1.4219, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -2.5598270893096924, | |
| "rewards/margins": 0.6573570370674133, | |
| "rewards/rejected": -3.217184543609619, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.6625701309110339, | |
| "grad_norm": 35.891043777341494, | |
| "learning_rate": 3.069319753571269e-07, | |
| "logits/chosen": -0.9791940450668335, | |
| "logits/rejected": -1.024137258529663, | |
| "logps/chosen": -1.0476067066192627, | |
| "logps/rejected": -1.3181588649749756, | |
| "loss": 1.4695, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.619016170501709, | |
| "rewards/margins": 0.6763805747032166, | |
| "rewards/rejected": -3.2953972816467285, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.673256745925728, | |
| "grad_norm": 22.292608497489727, | |
| "learning_rate": 2.898213858452173e-07, | |
| "logits/chosen": -1.1296494007110596, | |
| "logits/rejected": -1.086306095123291, | |
| "logps/chosen": -1.0372415781021118, | |
| "logps/rejected": -1.1678041219711304, | |
| "loss": 1.469, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.593104124069214, | |
| "rewards/margins": 0.3264063894748688, | |
| "rewards/rejected": -2.9195103645324707, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.6839433609404221, | |
| "grad_norm": 24.37127385244042, | |
| "learning_rate": 2.730047501302266e-07, | |
| "logits/chosen": -1.0386595726013184, | |
| "logits/rejected": -1.019863486289978, | |
| "logps/chosen": -0.9849531054496765, | |
| "logps/rejected": -1.240861177444458, | |
| "loss": 1.4399, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.4623827934265137, | |
| "rewards/margins": 0.6397703289985657, | |
| "rewards/rejected": -3.1021530628204346, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6946299759551162, | |
| "grad_norm": 28.3616490001571, | |
| "learning_rate": 2.5650558779781635e-07, | |
| "logits/chosen": -0.9190061688423157, | |
| "logits/rejected": -0.9376241564750671, | |
| "logps/chosen": -1.0179104804992676, | |
| "logps/rejected": -1.1567823886871338, | |
| "loss": 1.3944, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.544776439666748, | |
| "rewards/margins": 0.3471793532371521, | |
| "rewards/rejected": -2.891955614089966, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.7053165909698104, | |
| "grad_norm": 23.089017555469844, | |
| "learning_rate": 2.403469744184154e-07, | |
| "logits/chosen": -1.0184417963027954, | |
| "logits/rejected": -0.9639641642570496, | |
| "logps/chosen": -1.065198540687561, | |
| "logps/rejected": -1.4555224180221558, | |
| "loss": 1.4179, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -2.6629960536956787, | |
| "rewards/margins": 0.9758095741271973, | |
| "rewards/rejected": -3.638806104660034, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7160032059845044, | |
| "grad_norm": 19.01160663909916, | |
| "learning_rate": 2.2455150927394878e-07, | |
| "logits/chosen": -0.9782741665840149, | |
| "logits/rejected": -1.0168932676315308, | |
| "logps/chosen": -1.1860311031341553, | |
| "logps/rejected": -1.3966830968856812, | |
| "loss": 1.4107, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.965078115463257, | |
| "rewards/margins": 0.5266298651695251, | |
| "rewards/rejected": -3.4917080402374268, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.7266898209991985, | |
| "grad_norm": 24.023962985538827, | |
| "learning_rate": 2.0914128375069722e-07, | |
| "logits/chosen": -0.9762552380561829, | |
| "logits/rejected": -0.9652606248855591, | |
| "logps/chosen": -1.0335513353347778, | |
| "logps/rejected": -1.3794705867767334, | |
| "loss": 1.4363, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -2.5838780403137207, | |
| "rewards/margins": 0.8647986650466919, | |
| "rewards/rejected": -3.448676586151123, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7373764360138926, | |
| "grad_norm": 24.338376358621, | |
| "learning_rate": 1.9413785044249676e-07, | |
| "logits/chosen": -0.9369735717773438, | |
| "logits/rejected": -0.9039338827133179, | |
| "logps/chosen": -1.2081493139266968, | |
| "logps/rejected": -1.3959646224975586, | |
| "loss": 1.3412, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -3.020373821258545, | |
| "rewards/margins": 0.4695381224155426, | |
| "rewards/rejected": -3.4899115562438965, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.7480630510285867, | |
| "grad_norm": 25.7573169489427, | |
| "learning_rate": 1.7956219300748792e-07, | |
| "logits/chosen": -1.0008254051208496, | |
| "logits/rejected": -1.0084810256958008, | |
| "logps/chosen": -1.048879861831665, | |
| "logps/rejected": -1.3672925233840942, | |
| "loss": 1.3953, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -2.622199773788452, | |
| "rewards/margins": 0.7960314750671387, | |
| "rewards/rejected": -3.418231248855591, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7587496660432808, | |
| "grad_norm": 26.96889041092714, | |
| "learning_rate": 1.6543469682057104e-07, | |
| "logits/chosen": -1.0293567180633545, | |
| "logits/rejected": -0.9677215814590454, | |
| "logps/chosen": -1.0719913244247437, | |
| "logps/rejected": -1.2735936641693115, | |
| "loss": 1.3954, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.679978370666504, | |
| "rewards/margins": 0.5040060877799988, | |
| "rewards/rejected": -3.1839847564697266, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.7694362810579749, | |
| "grad_norm": 21.661999756036945, | |
| "learning_rate": 1.5177512046261666e-07, | |
| "logits/chosen": -0.9887057542800903, | |
| "logits/rejected": -1.0177589654922485, | |
| "logps/chosen": -1.291603922843933, | |
| "logps/rejected": -1.665748953819275, | |
| "loss": 1.3804, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -3.2290101051330566, | |
| "rewards/margins": 0.9353626370429993, | |
| "rewards/rejected": -4.164372444152832, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7801228960726689, | |
| "grad_norm": 21.098443523944745, | |
| "learning_rate": 1.3860256808630427e-07, | |
| "logits/chosen": -0.9724413752555847, | |
| "logits/rejected": -0.9351035952568054, | |
| "logps/chosen": -1.203754186630249, | |
| "logps/rejected": -1.3952935934066772, | |
| "loss": 1.3868, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -3.009385347366333, | |
| "rewards/margins": 0.4788486063480377, | |
| "rewards/rejected": -3.488234043121338, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.7908095110873631, | |
| "grad_norm": 25.021776262301575, | |
| "learning_rate": 1.2593546269723647e-07, | |
| "logits/chosen": -1.0209473371505737, | |
| "logits/rejected": -1.0015369653701782, | |
| "logps/chosen": -1.2007997035980225, | |
| "logps/rejected": -1.3749370574951172, | |
| "loss": 1.3739, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -3.0019993782043457, | |
| "rewards/margins": 0.43534326553344727, | |
| "rewards/rejected": -3.437342405319214, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8014961261020572, | |
| "grad_norm": 25.159927366829866, | |
| "learning_rate": 1.1379152038770029e-07, | |
| "logits/chosen": -1.0015761852264404, | |
| "logits/rejected": -1.0289709568023682, | |
| "logps/chosen": -1.0982341766357422, | |
| "logps/rejected": -1.3828232288360596, | |
| "loss": 1.4606, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -2.7455852031707764, | |
| "rewards/margins": 0.7114725708961487, | |
| "rewards/rejected": -3.4570579528808594, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 20.84984615780728, | |
| "learning_rate": 1.0218772555910954e-07, | |
| "logits/chosen": -1.087043046951294, | |
| "logits/rejected": -1.0346585512161255, | |
| "logps/chosen": -1.2293280363082886, | |
| "logps/rejected": -1.4728463888168335, | |
| "loss": 1.3777, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -3.073319911956787, | |
| "rewards/margins": 0.6087957620620728, | |
| "rewards/rejected": -3.6821160316467285, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8228693561314454, | |
| "grad_norm": 22.784469290812634, | |
| "learning_rate": 9.114030716778432e-08, | |
| "logits/chosen": -1.0404977798461914, | |
| "logits/rejected": -1.0313512086868286, | |
| "logps/chosen": -1.2078773975372314, | |
| "logps/rejected": -1.5215582847595215, | |
| "loss": 1.3372, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -3.019693613052368, | |
| "rewards/margins": 0.7842024564743042, | |
| "rewards/rejected": -3.803895950317383, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.8335559711461394, | |
| "grad_norm": 38.2013964652558, | |
| "learning_rate": 8.066471602728803e-08, | |
| "logits/chosen": -1.0984976291656494, | |
| "logits/rejected": -1.0727328062057495, | |
| "logps/chosen": -1.2440178394317627, | |
| "logps/rejected": -1.535946011543274, | |
| "loss": 1.3633, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -3.110044479370117, | |
| "rewards/margins": 0.7298205494880676, | |
| "rewards/rejected": -3.839865207672119, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8442425861608336, | |
| "grad_norm": 21.29430884924605, | |
| "learning_rate": 7.077560319906694e-08, | |
| "logits/chosen": -0.9021504521369934, | |
| "logits/rejected": -0.817115306854248, | |
| "logps/chosen": -1.1583576202392578, | |
| "logps/rejected": -1.3431838750839233, | |
| "loss": 1.396, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.8958938121795654, | |
| "rewards/margins": 0.46206584572792053, | |
| "rewards/rejected": -3.357959747314453, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.8549292011755276, | |
| "grad_norm": 18.542727738232298, | |
| "learning_rate": 6.148679950161672e-08, | |
| "logits/chosen": -0.9785951375961304, | |
| "logits/rejected": -0.9371121525764465, | |
| "logps/chosen": -1.3020581007003784, | |
| "logps/rejected": -1.5772645473480225, | |
| "loss": 1.3975, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -3.255145311355591, | |
| "rewards/margins": 0.6880159974098206, | |
| "rewards/rejected": -3.9431610107421875, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8549292011755276, | |
| "eval_logits/chosen": -1.1872848272323608, | |
| "eval_logits/rejected": -1.1544642448425293, | |
| "eval_logps/chosen": -1.1779358386993408, | |
| "eval_logps/rejected": -1.4415003061294556, | |
| "eval_loss": 1.375516653060913, | |
| "eval_rewards/accuracies": 0.6612903475761414, | |
| "eval_rewards/chosen": -2.9448394775390625, | |
| "eval_rewards/margins": 0.6589111685752869, | |
| "eval_rewards/rejected": -3.603750705718994, | |
| "eval_runtime": 72.7978, | |
| "eval_samples_per_second": 26.938, | |
| "eval_steps_per_second": 0.852, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8656158161902218, | |
| "grad_norm": 23.254725123506937, | |
| "learning_rate": 5.2811296166831666e-08, | |
| "logits/chosen": -0.9737190008163452, | |
| "logits/rejected": -0.8999165296554565, | |
| "logps/chosen": -1.2110763788223267, | |
| "logps/rejected": -1.4770066738128662, | |
| "loss": 1.4197, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -3.027691125869751, | |
| "rewards/margins": 0.6648265719413757, | |
| "rewards/rejected": -3.6925175189971924, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.8763024312049158, | |
| "grad_norm": 31.565832295867562, | |
| "learning_rate": 4.4761226670592066e-08, | |
| "logits/chosen": -0.9489310383796692, | |
| "logits/rejected": -0.9287575483322144, | |
| "logps/chosen": -1.2163779735565186, | |
| "logps/rejected": -1.4926642179489136, | |
| "loss": 1.3848, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -3.040944814682007, | |
| "rewards/margins": 0.6907154321670532, | |
| "rewards/rejected": -3.7316603660583496, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.88698904621961, | |
| "grad_norm": 22.344503337533215, | |
| "learning_rate": 3.734784976300165e-08, | |
| "logits/chosen": -0.92424076795578, | |
| "logits/rejected": -0.913791298866272, | |
| "logps/chosen": -1.1991280317306519, | |
| "logps/rejected": -1.3796842098236084, | |
| "loss": 1.3718, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.9978199005126953, | |
| "rewards/margins": 0.45139074325561523, | |
| "rewards/rejected": -3.4492106437683105, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.897675661234304, | |
| "grad_norm": 25.79666616123638, | |
| "learning_rate": 3.058153372200695e-08, | |
| "logits/chosen": -0.9958304166793823, | |
| "logits/rejected": -0.9945958256721497, | |
| "logps/chosen": -1.3022658824920654, | |
| "logps/rejected": -1.4703620672225952, | |
| "loss": 1.4021, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -3.255664348602295, | |
| "rewards/margins": 0.4202408194541931, | |
| "rewards/rejected": -3.675905704498291, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9083622762489981, | |
| "grad_norm": 24.252773163543072, | |
| "learning_rate": 2.4471741852423233e-08, | |
| "logits/chosen": -0.9615448713302612, | |
| "logits/rejected": -0.9239951968193054, | |
| "logps/chosen": -1.2393920421600342, | |
| "logps/rejected": -1.5796483755111694, | |
| "loss": 1.3444, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -3.098480224609375, | |
| "rewards/margins": 0.8506406545639038, | |
| "rewards/rejected": -3.9491209983825684, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.9190488912636923, | |
| "grad_norm": 25.840588676616843, | |
| "learning_rate": 1.9027019250647036e-08, | |
| "logits/chosen": -0.9318382143974304, | |
| "logits/rejected": -0.895746111869812, | |
| "logps/chosen": -1.19111168384552, | |
| "logps/rejected": -1.4055861234664917, | |
| "loss": 1.3383, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.9777793884277344, | |
| "rewards/margins": 0.536186158657074, | |
| "rewards/rejected": -3.513965606689453, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9297355062783863, | |
| "grad_norm": 24.38677858349731, | |
| "learning_rate": 1.4254980853566246e-08, | |
| "logits/chosen": -0.9664777517318726, | |
| "logits/rejected": -0.9452959299087524, | |
| "logps/chosen": -1.1789872646331787, | |
| "logps/rejected": -1.4990911483764648, | |
| "loss": 1.3959, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -2.9474682807922363, | |
| "rewards/margins": 0.8002597093582153, | |
| "rewards/rejected": -3.747727870941162, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.9404221212930804, | |
| "grad_norm": 20.846346077184034, | |
| "learning_rate": 1.016230078838226e-08, | |
| "logits/chosen": -0.9898012280464172, | |
| "logits/rejected": -0.9755558967590332, | |
| "logps/chosen": -1.205000638961792, | |
| "logps/rejected": -1.4142088890075684, | |
| "loss": 1.3646, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -3.0125012397766113, | |
| "rewards/margins": 0.5230205059051514, | |
| "rewards/rejected": -3.535521984100342, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9511087363077745, | |
| "grad_norm": 25.087392613196315, | |
| "learning_rate": 6.754703038239329e-09, | |
| "logits/chosen": -0.9521238207817078, | |
| "logits/rejected": -0.9080606698989868, | |
| "logps/chosen": -1.1550867557525635, | |
| "logps/rejected": -1.3055427074432373, | |
| "loss": 1.3825, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.887716770172119, | |
| "rewards/margins": 0.3761400282382965, | |
| "rewards/rejected": -3.263856887817383, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.9617953513224686, | |
| "grad_norm": 24.725896653899223, | |
| "learning_rate": 4.036953436716895e-09, | |
| "logits/chosen": -0.9332659840583801, | |
| "logits/rejected": -0.9464728236198425, | |
| "logps/chosen": -1.2779486179351807, | |
| "logps/rejected": -1.5308974981307983, | |
| "loss": 1.3408, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -3.194871425628662, | |
| "rewards/margins": 0.6323727369308472, | |
| "rewards/rejected": -3.8272438049316406, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9724819663371627, | |
| "grad_norm": 28.975077371611935, | |
| "learning_rate": 2.0128530023804656e-09, | |
| "logits/chosen": -0.9732203483581543, | |
| "logits/rejected": -0.9457721710205078, | |
| "logps/chosen": -1.345735788345337, | |
| "logps/rejected": -1.4330954551696777, | |
| "loss": 1.4589, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.3643393516540527, | |
| "rewards/margins": 0.21839912235736847, | |
| "rewards/rejected": -3.5827383995056152, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.9831685813518568, | |
| "grad_norm": 26.785266092328502, | |
| "learning_rate": 6.852326227130833e-10, | |
| "logits/chosen": -0.9476199150085449, | |
| "logits/rejected": -0.9895550608634949, | |
| "logps/chosen": -1.4385395050048828, | |
| "logps/rejected": -1.8012183904647827, | |
| "loss": 1.3759, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -3.596348524093628, | |
| "rewards/margins": 0.906697154045105, | |
| "rewards/rejected": -4.503045558929443, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.9938551963665508, | |
| "grad_norm": 25.28595563871724, | |
| "learning_rate": 5.594909486328348e-11, | |
| "logits/chosen": -1.0385969877243042, | |
| "logits/rejected": -0.9505317807197571, | |
| "logps/chosen": -1.1505335569381714, | |
| "logps/rejected": -1.4205496311187744, | |
| "loss": 1.3506, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.8763339519500732, | |
| "rewards/margins": 0.6750401258468628, | |
| "rewards/rejected": -3.5513739585876465, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.9981298423724285, | |
| "step": 467, | |
| "total_flos": 0.0, | |
| "train_loss": 1.4758259897824273, | |
| "train_runtime": 7808.6202, | |
| "train_samples_per_second": 7.668, | |
| "train_steps_per_second": 0.06 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 467, | |
| "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 | |
| } | |