| { | |
| "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": 8.463839594398154, | |
| "learning_rate": 1.0638297872340425e-08, | |
| "logits/chosen": -5.03189754486084, | |
| "logits/rejected": -4.649786472320557, | |
| "logps/chosen": -1.325859785079956, | |
| "logps/rejected": -1.2691845893859863, | |
| "loss": 1.6361, | |
| "rewards/accuracies": 0.375, | |
| "rewards/chosen": -3.3146491050720215, | |
| "rewards/margins": -0.1416877955198288, | |
| "rewards/rejected": -3.172961473464966, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010686615014694095, | |
| "grad_norm": 10.461652514268042, | |
| "learning_rate": 5.3191489361702123e-08, | |
| "logits/chosen": -4.613029479980469, | |
| "logits/rejected": -4.432791709899902, | |
| "logps/chosen": -1.3983737230300903, | |
| "logps/rejected": -1.4233554601669312, | |
| "loss": 1.8223, | |
| "rewards/accuracies": 0.453125, | |
| "rewards/chosen": -3.495934247970581, | |
| "rewards/margins": 0.0624544657766819, | |
| "rewards/rejected": -3.5583884716033936, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02137323002938819, | |
| "grad_norm": 14.978344124531132, | |
| "learning_rate": 1.0638297872340425e-07, | |
| "logits/chosen": -4.634655952453613, | |
| "logits/rejected": -4.519498348236084, | |
| "logps/chosen": -1.5365169048309326, | |
| "logps/rejected": -1.5281957387924194, | |
| "loss": 1.8105, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.8412926197052, | |
| "rewards/margins": -0.020802771672606468, | |
| "rewards/rejected": -3.8204894065856934, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03205984504408229, | |
| "grad_norm": 9.265143341850845, | |
| "learning_rate": 1.5957446808510638e-07, | |
| "logits/chosen": -4.536217212677002, | |
| "logits/rejected": -4.47308874130249, | |
| "logps/chosen": -1.5140950679779053, | |
| "logps/rejected": -1.4939788579940796, | |
| "loss": 1.748, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.785238265991211, | |
| "rewards/margins": -0.05029074102640152, | |
| "rewards/rejected": -3.7349472045898438, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04274646005877638, | |
| "grad_norm": 7.20742555554014, | |
| "learning_rate": 2.127659574468085e-07, | |
| "logits/chosen": -4.679377555847168, | |
| "logits/rejected": -4.679813385009766, | |
| "logps/chosen": -1.4990894794464111, | |
| "logps/rejected": -1.5615761280059814, | |
| "loss": 1.7569, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.7477238178253174, | |
| "rewards/margins": 0.15621671080589294, | |
| "rewards/rejected": -3.9039406776428223, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.053433075073470476, | |
| "grad_norm": 8.314479583512021, | |
| "learning_rate": 2.659574468085106e-07, | |
| "logits/chosen": -4.735424041748047, | |
| "logits/rejected": -4.67797327041626, | |
| "logps/chosen": -1.5032179355621338, | |
| "logps/rejected": -1.5036437511444092, | |
| "loss": 1.7663, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.758044719696045, | |
| "rewards/margins": 0.001064938260242343, | |
| "rewards/rejected": -3.7591099739074707, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06411969008816458, | |
| "grad_norm": 8.24895452301186, | |
| "learning_rate": 3.1914893617021275e-07, | |
| "logits/chosen": -4.979925632476807, | |
| "logits/rejected": -4.7350873947143555, | |
| "logps/chosen": -1.526336431503296, | |
| "logps/rejected": -1.5037505626678467, | |
| "loss": 1.7647, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.81584095954895, | |
| "rewards/margins": -0.05646459385752678, | |
| "rewards/rejected": -3.7593765258789062, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07480630510285867, | |
| "grad_norm": 8.218009051807567, | |
| "learning_rate": 3.7234042553191484e-07, | |
| "logits/chosen": -4.755112171173096, | |
| "logits/rejected": -4.576828479766846, | |
| "logps/chosen": -1.5506718158721924, | |
| "logps/rejected": -1.5067684650421143, | |
| "loss": 1.7863, | |
| "rewards/accuracies": 0.32499998807907104, | |
| "rewards/chosen": -3.876680374145508, | |
| "rewards/margins": -0.10975880920886993, | |
| "rewards/rejected": -3.7669215202331543, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08549292011755276, | |
| "grad_norm": 10.346763541049816, | |
| "learning_rate": 4.25531914893617e-07, | |
| "logits/chosen": -4.490562915802002, | |
| "logits/rejected": -4.430506706237793, | |
| "logps/chosen": -1.4821277856826782, | |
| "logps/rejected": -1.5026350021362305, | |
| "loss": 1.7765, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -3.7053191661834717, | |
| "rewards/margins": 0.05126795917749405, | |
| "rewards/rejected": -3.756587266921997, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09617953513224686, | |
| "grad_norm": 8.159954026995056, | |
| "learning_rate": 4.787234042553192e-07, | |
| "logits/chosen": -4.584813594818115, | |
| "logits/rejected": -4.397616386413574, | |
| "logps/chosen": -1.5396957397460938, | |
| "logps/rejected": -1.5210858583450317, | |
| "loss": 1.7353, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.8492393493652344, | |
| "rewards/margins": -0.046524763107299805, | |
| "rewards/rejected": -3.8027145862579346, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10686615014694095, | |
| "grad_norm": 6.871700800549116, | |
| "learning_rate": 4.999370587356267e-07, | |
| "logits/chosen": -4.505709648132324, | |
| "logits/rejected": -4.605370998382568, | |
| "logps/chosen": -1.4193190336227417, | |
| "logps/rejected": -1.5022988319396973, | |
| "loss": 1.7108, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.54829740524292, | |
| "rewards/margins": 0.20744964480400085, | |
| "rewards/rejected": -3.7557475566864014, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11755276516163506, | |
| "grad_norm": 8.013382748882483, | |
| "learning_rate": 4.995525324419337e-07, | |
| "logits/chosen": -4.650801658630371, | |
| "logits/rejected": -4.462812900543213, | |
| "logps/chosen": -1.3039531707763672, | |
| "logps/rejected": -1.3392069339752197, | |
| "loss": 1.7299, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -3.259882688522339, | |
| "rewards/margins": 0.08813469856977463, | |
| "rewards/rejected": -3.348017454147339, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.12823938017632916, | |
| "grad_norm": 8.303134763756917, | |
| "learning_rate": 4.988189843662815e-07, | |
| "logits/chosen": -4.751210689544678, | |
| "logits/rejected": -4.686429023742676, | |
| "logps/chosen": -1.2432037591934204, | |
| "logps/rejected": -1.2768182754516602, | |
| "loss": 1.7359, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -3.1080093383789062, | |
| "rewards/margins": 0.08403660356998444, | |
| "rewards/rejected": -3.1920459270477295, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13892599519102325, | |
| "grad_norm": 6.777024673739444, | |
| "learning_rate": 4.977374404419837e-07, | |
| "logits/chosen": -4.5330071449279785, | |
| "logits/rejected": -4.71773099899292, | |
| "logps/chosen": -1.3066389560699463, | |
| "logps/rejected": -1.2977256774902344, | |
| "loss": 1.7457, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.266597270965576, | |
| "rewards/margins": -0.02228318527340889, | |
| "rewards/rejected": -3.244314193725586, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.14961261020571734, | |
| "grad_norm": 9.380383365435003, | |
| "learning_rate": 4.963094133060148e-07, | |
| "logits/chosen": -4.556839466094971, | |
| "logits/rejected": -4.482637882232666, | |
| "logps/chosen": -1.3503261804580688, | |
| "logps/rejected": -1.309463620185852, | |
| "loss": 1.7783, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -3.3758158683776855, | |
| "rewards/margins": -0.1021568551659584, | |
| "rewards/rejected": -3.2736592292785645, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.16029922522041143, | |
| "grad_norm": 8.316182420752773, | |
| "learning_rate": 4.945369001834514e-07, | |
| "logits/chosen": -4.512010097503662, | |
| "logits/rejected": -4.473996639251709, | |
| "logps/chosen": -1.3320636749267578, | |
| "logps/rejected": -1.3820534944534302, | |
| "loss": 1.6619, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.3301587104797363, | |
| "rewards/margins": 0.12497501075267792, | |
| "rewards/rejected": -3.455134153366089, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.17098584023510552, | |
| "grad_norm": 7.968870017741486, | |
| "learning_rate": 4.924223800941717e-07, | |
| "logits/chosen": -4.695282936096191, | |
| "logits/rejected": -4.676292419433594, | |
| "logps/chosen": -1.3696571588516235, | |
| "logps/rejected": -1.341069221496582, | |
| "loss": 1.7283, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.4241433143615723, | |
| "rewards/margins": -0.0714699923992157, | |
| "rewards/rejected": -3.352673053741455, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.18167245524979964, | |
| "grad_norm": 6.985848218948992, | |
| "learning_rate": 4.899688103857222e-07, | |
| "logits/chosen": -4.516369342803955, | |
| "logits/rejected": -4.401045799255371, | |
| "logps/chosen": -1.3444452285766602, | |
| "logps/rejected": -1.356410264968872, | |
| "loss": 1.7358, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.3611133098602295, | |
| "rewards/margins": 0.02991255186498165, | |
| "rewards/rejected": -3.3910255432128906, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.19235907026449373, | |
| "grad_norm": 7.509298977705249, | |
| "learning_rate": 4.871796225971999e-07, | |
| "logits/chosen": -4.371309280395508, | |
| "logits/rejected": -4.230607509613037, | |
| "logps/chosen": -1.4209774732589722, | |
| "logps/rejected": -1.3977888822555542, | |
| "loss": 1.7377, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.552443265914917, | |
| "rewards/margins": -0.05797123908996582, | |
| "rewards/rejected": -3.494471788406372, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 7.494464457003994, | |
| "learning_rate": 4.840587176599343e-07, | |
| "logits/chosen": -4.4767680168151855, | |
| "logits/rejected": -4.559601306915283, | |
| "logps/chosen": -1.3669825792312622, | |
| "logps/rejected": -1.4395772218704224, | |
| "loss": 1.7103, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -3.4174563884735107, | |
| "rewards/margins": 0.181486576795578, | |
| "rewards/rejected": -3.598942995071411, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.2137323002938819, | |
| "grad_norm": 9.417753270846694, | |
| "learning_rate": 4.806104604416823e-07, | |
| "logits/chosen": -4.715841770172119, | |
| "logits/rejected": -4.59860897064209, | |
| "logps/chosen": -1.341230869293213, | |
| "logps/rejected": -1.2855507135391235, | |
| "loss": 1.7501, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.3530774116516113, | |
| "rewards/margins": -0.13920024037361145, | |
| "rewards/rejected": -3.2138772010803223, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.224418915308576, | |
| "grad_norm": 8.25939695647074, | |
| "learning_rate": 4.768396736419662e-07, | |
| "logits/chosen": -4.579805850982666, | |
| "logits/rejected": -4.472721576690674, | |
| "logps/chosen": -1.2403936386108398, | |
| "logps/rejected": -1.2722570896148682, | |
| "loss": 1.7103, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.1009843349456787, | |
| "rewards/margins": 0.07965835183858871, | |
| "rewards/rejected": -3.180642604827881, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.2351055303232701, | |
| "grad_norm": 7.0286538282806905, | |
| "learning_rate": 4.7275163104709194e-07, | |
| "logits/chosen": -4.310456275939941, | |
| "logits/rejected": -4.180710315704346, | |
| "logps/chosen": -1.3125991821289062, | |
| "logps/rejected": -1.3054355382919312, | |
| "loss": 1.7814, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.2814979553222656, | |
| "rewards/margins": -0.017909258604049683, | |
| "rewards/rejected": -3.2635891437530518, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2457921453379642, | |
| "grad_norm": 7.4251876749652315, | |
| "learning_rate": 4.683520501542824e-07, | |
| "logits/chosen": -4.345327854156494, | |
| "logits/rejected": -4.324080944061279, | |
| "logps/chosen": -1.269272804260254, | |
| "logps/rejected": -1.255380392074585, | |
| "loss": 1.7077, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.1731820106506348, | |
| "rewards/margins": -0.03473084047436714, | |
| "rewards/rejected": -3.138451099395752, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.2564787603526583, | |
| "grad_norm": 7.68817663348518, | |
| "learning_rate": 4.636470841752404e-07, | |
| "logits/chosen": -4.619086265563965, | |
| "logits/rejected": -4.4949049949646, | |
| "logps/chosen": -1.2920095920562744, | |
| "logps/rejected": -1.2791041135787964, | |
| "loss": 1.7086, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -3.2300238609313965, | |
| "rewards/margins": -0.03226359933614731, | |
| "rewards/rejected": -3.1977603435516357, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2671653753673524, | |
| "grad_norm": 7.125331982786664, | |
| "learning_rate": 4.5864331343032565e-07, | |
| "logits/chosen": -4.59976053237915, | |
| "logits/rejected": -4.594804286956787, | |
| "logps/chosen": -1.3433685302734375, | |
| "logps/rejected": -1.4002816677093506, | |
| "loss": 1.729, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -3.3584213256835938, | |
| "rewards/margins": 0.14228278398513794, | |
| "rewards/rejected": -3.500704288482666, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2778519903820465, | |
| "grad_norm": 8.312615727941836, | |
| "learning_rate": 4.533477361453819e-07, | |
| "logits/chosen": -4.355613708496094, | |
| "logits/rejected": -4.472823619842529, | |
| "logps/chosen": -1.3405156135559082, | |
| "logps/rejected": -1.3726789951324463, | |
| "loss": 1.7552, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.3512885570526123, | |
| "rewards/margins": 0.08040888607501984, | |
| "rewards/rejected": -3.431697368621826, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2885386053967406, | |
| "grad_norm": 8.624812004095546, | |
| "learning_rate": 4.4776775866408533e-07, | |
| "logits/chosen": -4.605002403259277, | |
| "logits/rejected": -4.486802577972412, | |
| "logps/chosen": -1.2775276899337769, | |
| "logps/rejected": -1.2938398122787476, | |
| "loss": 1.7193, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -3.193819284439087, | |
| "rewards/margins": 0.040780358016490936, | |
| "rewards/rejected": -3.2345993518829346, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.2992252204114347, | |
| "grad_norm": 8.928849598835027, | |
| "learning_rate": 4.4191118508950277e-07, | |
| "logits/chosen": -4.492905616760254, | |
| "logits/rejected": -4.466243743896484, | |
| "logps/chosen": -1.4048876762390137, | |
| "logps/rejected": -1.4387624263763428, | |
| "loss": 1.7204, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -3.512219190597534, | |
| "rewards/margins": 0.08468694984912872, | |
| "rewards/rejected": -3.5969061851501465, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.30991183542612877, | |
| "grad_norm": 8.071911039943062, | |
| "learning_rate": 4.357862063693485e-07, | |
| "logits/chosen": -4.527454376220703, | |
| "logits/rejected": -4.5367279052734375, | |
| "logps/chosen": -1.344947099685669, | |
| "logps/rejected": -1.371901273727417, | |
| "loss": 1.7418, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.362368106842041, | |
| "rewards/margins": 0.06738501042127609, | |
| "rewards/rejected": -3.429752826690674, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.32059845044082286, | |
| "grad_norm": 8.671271998589708, | |
| "learning_rate": 4.294013888402029e-07, | |
| "logits/chosen": -4.329895496368408, | |
| "logits/rejected": -4.347752571105957, | |
| "logps/chosen": -1.2269880771636963, | |
| "logps/rejected": -1.3332128524780273, | |
| "loss": 1.7002, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -3.0674703121185303, | |
| "rewards/margins": 0.2655617594718933, | |
| "rewards/rejected": -3.3330321311950684, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33128506545551695, | |
| "grad_norm": 11.410050720477157, | |
| "learning_rate": 4.227656622467162e-07, | |
| "logits/chosen": -4.60593843460083, | |
| "logits/rejected": -4.5206708908081055, | |
| "logps/chosen": -1.2302855253219604, | |
| "logps/rejected": -1.285362958908081, | |
| "loss": 1.7276, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.075713872909546, | |
| "rewards/margins": 0.13769367337226868, | |
| "rewards/rejected": -3.2134079933166504, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.34197168047021104, | |
| "grad_norm": 8.929425526594324, | |
| "learning_rate": 4.158883072525528e-07, | |
| "logits/chosen": -4.636877536773682, | |
| "logits/rejected": -4.64215612411499, | |
| "logps/chosen": -1.2857083082199097, | |
| "logps/rejected": -1.2909152507781982, | |
| "loss": 1.7261, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.214270830154419, | |
| "rewards/margins": 0.01301727257668972, | |
| "rewards/rejected": -3.227288007736206, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3526582954849052, | |
| "grad_norm": 8.999759770579269, | |
| "learning_rate": 4.087789424605447e-07, | |
| "logits/chosen": -4.477012634277344, | |
| "logits/rejected": -4.342657089233398, | |
| "logps/chosen": -1.2676000595092773, | |
| "logps/rejected": -1.2652432918548584, | |
| "loss": 1.7307, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -3.168999671936035, | |
| "rewards/margins": -0.005891731474548578, | |
| "rewards/rejected": -3.1631083488464355, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.36334491049959927, | |
| "grad_norm": 7.38370964011058, | |
| "learning_rate": 4.0144751096020497e-07, | |
| "logits/chosen": -4.322202682495117, | |
| "logits/rejected": -4.344858646392822, | |
| "logps/chosen": -1.3334827423095703, | |
| "logps/rejected": -1.3564600944519043, | |
| "loss": 1.7042, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.3337066173553467, | |
| "rewards/margins": 0.05744355916976929, | |
| "rewards/rejected": -3.3911502361297607, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.37403152551429336, | |
| "grad_norm": 7.609720560260437, | |
| "learning_rate": 3.939042664214184e-07, | |
| "logits/chosen": -4.588566780090332, | |
| "logits/rejected": -4.577520847320557, | |
| "logps/chosen": -1.3666011095046997, | |
| "logps/rejected": -1.4189504384994507, | |
| "loss": 1.6971, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -3.4165027141571045, | |
| "rewards/margins": 0.13087329268455505, | |
| "rewards/rejected": -3.5473761558532715, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.38471814052898745, | |
| "grad_norm": 7.522883946304073, | |
| "learning_rate": 3.8615975875375676e-07, | |
| "logits/chosen": -4.611153602600098, | |
| "logits/rejected": -4.639096736907959, | |
| "logps/chosen": -1.3451780080795288, | |
| "logps/rejected": -1.407920241355896, | |
| "loss": 1.7286, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -3.3629448413848877, | |
| "rewards/margins": 0.15685562789440155, | |
| "rewards/rejected": -3.5198006629943848, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.39540475554368154, | |
| "grad_norm": 8.15961865987409, | |
| "learning_rate": 3.7822481935147655e-07, | |
| "logits/chosen": -4.689408302307129, | |
| "logits/rejected": -4.5546979904174805, | |
| "logps/chosen": -1.3114018440246582, | |
| "logps/rejected": -1.333992838859558, | |
| "loss": 1.7275, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.2785046100616455, | |
| "rewards/margins": 0.05647749826312065, | |
| "rewards/rejected": -3.334981918334961, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 8.427441389080544, | |
| "learning_rate": 3.7011054594483443e-07, | |
| "logits/chosen": -4.588101863861084, | |
| "logits/rejected": -4.52250862121582, | |
| "logps/chosen": -1.3444761037826538, | |
| "logps/rejected": -1.4100086688995361, | |
| "loss": 1.6814, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -3.3611907958984375, | |
| "rewards/margins": 0.16383083164691925, | |
| "rewards/rejected": -3.5250213146209717, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4167779855730697, | |
| "grad_norm": 8.693634103691119, | |
| "learning_rate": 3.618282870789081e-07, | |
| "logits/chosen": -4.41694450378418, | |
| "logits/rejected": -4.407210826873779, | |
| "logps/chosen": -1.3954448699951172, | |
| "logps/rejected": -1.3883302211761475, | |
| "loss": 1.7438, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -3.488612413406372, | |
| "rewards/margins": -0.017786988988518715, | |
| "rewards/rejected": -3.4708251953125, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.4274646005877638, | |
| "grad_norm": 7.817917194646963, | |
| "learning_rate": 3.5338962624163016e-07, | |
| "logits/chosen": -4.41127872467041, | |
| "logits/rejected": -4.464591026306152, | |
| "logps/chosen": -1.3187341690063477, | |
| "logps/rejected": -1.3379795551300049, | |
| "loss": 1.7493, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.296835422515869, | |
| "rewards/margins": 0.04811403155326843, | |
| "rewards/rejected": -3.344949245452881, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4381512156024579, | |
| "grad_norm": 7.87816681095113, | |
| "learning_rate": 3.448063656632321e-07, | |
| "logits/chosen": -4.585541248321533, | |
| "logits/rejected": -4.440451145172119, | |
| "logps/chosen": -1.3579752445220947, | |
| "logps/rejected": -1.3495149612426758, | |
| "loss": 1.6908, | |
| "rewards/accuracies": 0.375, | |
| "rewards/chosen": -3.3949382305145264, | |
| "rewards/margins": -0.02115105651319027, | |
| "rewards/rejected": -3.3737874031066895, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.448837830617152, | |
| "grad_norm": 8.192581867844957, | |
| "learning_rate": 3.360905098097587e-07, | |
| "logits/chosen": -4.702603340148926, | |
| "logits/rejected": -4.5605058670043945, | |
| "logps/chosen": -1.3157254457473755, | |
| "logps/rejected": -1.326529860496521, | |
| "loss": 1.7765, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.289313554763794, | |
| "rewards/margins": 0.02701149880886078, | |
| "rewards/rejected": -3.3163249492645264, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.45952444563184613, | |
| "grad_norm": 9.916408251919977, | |
| "learning_rate": 3.272542485937368e-07, | |
| "logits/chosen": -4.547693252563477, | |
| "logits/rejected": -4.488272666931152, | |
| "logps/chosen": -1.2768534421920776, | |
| "logps/rejected": -1.3246023654937744, | |
| "loss": 1.6999, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -3.192133665084839, | |
| "rewards/margins": 0.11937196552753448, | |
| "rewards/rejected": -3.3115055561065674, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.4702110606465402, | |
| "grad_norm": 11.54297063214211, | |
| "learning_rate": 3.1830994032548e-07, | |
| "logits/chosen": -4.549879550933838, | |
| "logits/rejected": -4.424410820007324, | |
| "logps/chosen": -1.354941725730896, | |
| "logps/rejected": -1.3328911066055298, | |
| "loss": 1.7459, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -3.387354612350464, | |
| "rewards/margins": -0.05512649938464165, | |
| "rewards/rejected": -3.332228183746338, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.4808976756612343, | |
| "grad_norm": 7.745936948551203, | |
| "learning_rate": 3.0927009442887437e-07, | |
| "logits/chosen": -4.522200584411621, | |
| "logits/rejected": -4.433478355407715, | |
| "logps/chosen": -1.3177942037582397, | |
| "logps/rejected": -1.312595248222351, | |
| "loss": 1.7151, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -3.294485569000244, | |
| "rewards/margins": -0.012997297570109367, | |
| "rewards/rejected": -3.2814879417419434, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4915842906759284, | |
| "grad_norm": 8.982993035215612, | |
| "learning_rate": 3.001473539458182e-07, | |
| "logits/chosen": -4.790830135345459, | |
| "logits/rejected": -4.502416610717773, | |
| "logps/chosen": -1.2848570346832275, | |
| "logps/rejected": -1.3198983669281006, | |
| "loss": 1.7007, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.2121429443359375, | |
| "rewards/margins": 0.08760317414999008, | |
| "rewards/rejected": -3.299746036529541, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5022709056906225, | |
| "grad_norm": 7.9015489492124225, | |
| "learning_rate": 2.909544778537844e-07, | |
| "logits/chosen": -4.630471229553223, | |
| "logits/rejected": -4.5115509033203125, | |
| "logps/chosen": -1.3522698879241943, | |
| "logps/rejected": -1.337274193763733, | |
| "loss": 1.7185, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -3.3806750774383545, | |
| "rewards/margins": -0.03748970106244087, | |
| "rewards/rejected": -3.3431854248046875, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.5129575207053166, | |
| "grad_norm": 8.651538383776115, | |
| "learning_rate": 2.817043232212371e-07, | |
| "logits/chosen": -4.58451509475708, | |
| "logits/rejected": -4.510278701782227, | |
| "logps/chosen": -1.3096239566802979, | |
| "logps/rejected": -1.3383967876434326, | |
| "loss": 1.6684, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.274059295654297, | |
| "rewards/margins": 0.07193277776241302, | |
| "rewards/rejected": -3.34599232673645, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5236441357200107, | |
| "grad_norm": 8.276564878362981, | |
| "learning_rate": 2.7240982722585837e-07, | |
| "logits/chosen": -4.700469970703125, | |
| "logits/rejected": -4.604641914367676, | |
| "logps/chosen": -1.296589970588684, | |
| "logps/rejected": -1.295531988143921, | |
| "loss": 1.7276, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.2414746284484863, | |
| "rewards/margins": -0.0026448846329003572, | |
| "rewards/rejected": -3.2388298511505127, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.5343307507347048, | |
| "grad_norm": 8.739959615068186, | |
| "learning_rate": 2.63083989060736e-07, | |
| "logits/chosen": -4.688320636749268, | |
| "logits/rejected": -4.622137069702148, | |
| "logps/chosen": -1.420803427696228, | |
| "logps/rejected": -1.5039843320846558, | |
| "loss": 1.706, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.552008867263794, | |
| "rewards/margins": 0.20795249938964844, | |
| "rewards/rejected": -3.7599613666534424, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5450173657493989, | |
| "grad_norm": 8.205463032414816, | |
| "learning_rate": 2.537398517538159e-07, | |
| "logits/chosen": -4.498805046081543, | |
| "logits/rejected": -4.545159816741943, | |
| "logps/chosen": -1.3838533163070679, | |
| "logps/rejected": -1.427234411239624, | |
| "loss": 1.7113, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.4596335887908936, | |
| "rewards/margins": 0.10845281928777695, | |
| "rewards/rejected": -3.5680861473083496, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.555703980764093, | |
| "grad_norm": 9.403793469995996, | |
| "learning_rate": 2.4439048392604877e-07, | |
| "logits/chosen": -4.465289115905762, | |
| "logits/rejected": -4.415510654449463, | |
| "logps/chosen": -1.3352388143539429, | |
| "logps/rejected": -1.4621715545654297, | |
| "loss": 1.6556, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -3.338097095489502, | |
| "rewards/margins": 0.3173316717147827, | |
| "rewards/rejected": -3.655428647994995, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.566390595778787, | |
| "grad_norm": 9.87099892276829, | |
| "learning_rate": 2.3504896151374144e-07, | |
| "logits/chosen": -4.503221035003662, | |
| "logits/rejected": -4.532763481140137, | |
| "logps/chosen": -1.3822702169418335, | |
| "logps/rejected": -1.3944904804229736, | |
| "loss": 1.7068, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.4556756019592285, | |
| "rewards/margins": 0.03055066429078579, | |
| "rewards/rejected": -3.4862258434295654, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.5770772107934812, | |
| "grad_norm": 9.709835641918701, | |
| "learning_rate": 2.2572834948067795e-07, | |
| "logits/chosen": -4.5645036697387695, | |
| "logits/rejected": -4.5933027267456055, | |
| "logps/chosen": -1.2755674123764038, | |
| "logps/rejected": -1.3178540468215942, | |
| "loss": 1.6978, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.1889185905456543, | |
| "rewards/margins": 0.10571658611297607, | |
| "rewards/rejected": -3.294635057449341, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5877638258081752, | |
| "grad_norm": 10.023847803695947, | |
| "learning_rate": 2.164416835455862e-07, | |
| "logits/chosen": -4.247835159301758, | |
| "logits/rejected": -4.194937229156494, | |
| "logps/chosen": -1.3332767486572266, | |
| "logps/rejected": -1.3482983112335205, | |
| "loss": 1.6989, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.3331921100616455, | |
| "rewards/margins": 0.03755370154976845, | |
| "rewards/rejected": -3.3707454204559326, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5984504408228694, | |
| "grad_norm": 12.08817520180493, | |
| "learning_rate": 2.072019519505062e-07, | |
| "logits/chosen": -4.784656524658203, | |
| "logits/rejected": -4.656561851501465, | |
| "logps/chosen": -1.3093583583831787, | |
| "logps/rejected": -1.3450324535369873, | |
| "loss": 1.6805, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.2733960151672363, | |
| "rewards/margins": 0.08918508142232895, | |
| "rewards/rejected": -3.362581253051758, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 8.794437977214896, | |
| "learning_rate": 1.980220772955602e-07, | |
| "logits/chosen": -4.4688849449157715, | |
| "logits/rejected": -4.370944499969482, | |
| "logps/chosen": -1.335540533065796, | |
| "logps/rejected": -1.3469078540802002, | |
| "loss": 1.6982, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.3388514518737793, | |
| "rewards/margins": 0.028418350964784622, | |
| "rewards/rejected": -3.367269992828369, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.6198236708522575, | |
| "grad_norm": 13.055246394354226, | |
| "learning_rate": 1.8891489846552644e-07, | |
| "logits/chosen": -4.446020603179932, | |
| "logits/rejected": -4.3907389640808105, | |
| "logps/chosen": -1.332921028137207, | |
| "logps/rejected": -1.3605272769927979, | |
| "loss": 1.646, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.3323025703430176, | |
| "rewards/margins": 0.06901548057794571, | |
| "rewards/rejected": -3.401318311691284, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6305102858669517, | |
| "grad_norm": 9.19761267983714, | |
| "learning_rate": 1.7989315267349933e-07, | |
| "logits/chosen": -4.640501022338867, | |
| "logits/rejected": -4.582611083984375, | |
| "logps/chosen": -1.345268964767456, | |
| "logps/rejected": -1.3664015531539917, | |
| "loss": 1.6828, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -3.363173007965088, | |
| "rewards/margins": 0.05283108353614807, | |
| "rewards/rejected": -3.416003704071045, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.6411969008816457, | |
| "grad_norm": 8.163371854491718, | |
| "learning_rate": 1.7096945764674398e-07, | |
| "logits/chosen": -4.53090763092041, | |
| "logits/rejected": -4.449822425842285, | |
| "logps/chosen": -1.357412338256836, | |
| "logps/rejected": -1.340632677078247, | |
| "loss": 1.7361, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -3.393531084060669, | |
| "rewards/margins": -0.04194915294647217, | |
| "rewards/rejected": -3.3515820503234863, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6518835158963399, | |
| "grad_norm": 9.037247969057203, | |
| "learning_rate": 1.621562939796643e-07, | |
| "logits/chosen": -4.495213508605957, | |
| "logits/rejected": -4.3966827392578125, | |
| "logps/chosen": -1.4379794597625732, | |
| "logps/rejected": -1.473179578781128, | |
| "loss": 1.7018, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.5949490070343018, | |
| "rewards/margins": 0.08800011873245239, | |
| "rewards/rejected": -3.682948589324951, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.6625701309110339, | |
| "grad_norm": 13.030910514110566, | |
| "learning_rate": 1.5346598767856345e-07, | |
| "logits/chosen": -4.379870414733887, | |
| "logits/rejected": -4.560307025909424, | |
| "logps/chosen": -1.398297905921936, | |
| "logps/rejected": -1.4838316440582275, | |
| "loss": 1.6849, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -3.4957451820373535, | |
| "rewards/margins": 0.21383419632911682, | |
| "rewards/rejected": -3.7095787525177, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.673256745925728, | |
| "grad_norm": 9.601996227444385, | |
| "learning_rate": 1.4491069292260866e-07, | |
| "logits/chosen": -4.930965423583984, | |
| "logits/rejected": -4.876108646392822, | |
| "logps/chosen": -1.430011510848999, | |
| "logps/rejected": -1.4526395797729492, | |
| "loss": 1.7084, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.575028896331787, | |
| "rewards/margins": 0.05657036229968071, | |
| "rewards/rejected": -3.631598949432373, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.6839433609404221, | |
| "grad_norm": 8.370911393375984, | |
| "learning_rate": 1.365023750651133e-07, | |
| "logits/chosen": -4.7936224937438965, | |
| "logits/rejected": -4.740353584289551, | |
| "logps/chosen": -1.282147765159607, | |
| "logps/rejected": -1.4656821489334106, | |
| "loss": 1.6538, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -3.205369472503662, | |
| "rewards/margins": 0.4588359296321869, | |
| "rewards/rejected": -3.6642050743103027, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6946299759551162, | |
| "grad_norm": 10.059762786625233, | |
| "learning_rate": 1.2825279389890818e-07, | |
| "logits/chosen": -4.382018089294434, | |
| "logits/rejected": -4.410761833190918, | |
| "logps/chosen": -1.3325350284576416, | |
| "logps/rejected": -1.4160031080245972, | |
| "loss": 1.6493, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -3.3313376903533936, | |
| "rewards/margins": 0.20867028832435608, | |
| "rewards/rejected": -3.5400078296661377, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.7053165909698104, | |
| "grad_norm": 9.161029493778742, | |
| "learning_rate": 1.201734872092077e-07, | |
| "logits/chosen": -4.73400354385376, | |
| "logits/rejected": -4.58245849609375, | |
| "logps/chosen": -1.3593249320983887, | |
| "logps/rejected": -1.5017703771591187, | |
| "loss": 1.6762, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.3983123302459717, | |
| "rewards/margins": 0.3561137616634369, | |
| "rewards/rejected": -3.7544262409210205, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7160032059845044, | |
| "grad_norm": 8.300761837819268, | |
| "learning_rate": 1.1227575463697439e-07, | |
| "logits/chosen": -4.566507816314697, | |
| "logits/rejected": -4.648960590362549, | |
| "logps/chosen": -1.3698114156723022, | |
| "logps/rejected": -1.4295297861099243, | |
| "loss": 1.683, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.4245285987854004, | |
| "rewards/margins": 0.14929580688476562, | |
| "rewards/rejected": -3.573824405670166, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.7266898209991985, | |
| "grad_norm": 9.524750095557723, | |
| "learning_rate": 1.0457064187534861e-07, | |
| "logits/chosen": -4.421561241149902, | |
| "logits/rejected": -4.395345687866211, | |
| "logps/chosen": -1.2686647176742554, | |
| "logps/rejected": -1.3926507234573364, | |
| "loss": 1.7023, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -3.171661615371704, | |
| "rewards/margins": 0.3099651038646698, | |
| "rewards/rejected": -3.4816269874572754, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7373764360138926, | |
| "grad_norm": 9.264218645867375, | |
| "learning_rate": 9.706892522124838e-08, | |
| "logits/chosen": -4.445697784423828, | |
| "logits/rejected": -4.289374351501465, | |
| "logps/chosen": -1.3950523138046265, | |
| "logps/rejected": -1.3890727758407593, | |
| "loss": 1.6517, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -3.487631320953369, | |
| "rewards/margins": -0.014949033036828041, | |
| "rewards/rejected": -3.472681760787964, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.7480630510285867, | |
| "grad_norm": 9.805135028816036, | |
| "learning_rate": 8.978109650374396e-08, | |
| "logits/chosen": -4.54650354385376, | |
| "logits/rejected": -4.534266471862793, | |
| "logps/chosen": -1.4010884761810303, | |
| "logps/rejected": -1.4561593532562256, | |
| "loss": 1.7054, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -3.5027213096618652, | |
| "rewards/margins": 0.13767699897289276, | |
| "rewards/rejected": -3.6403985023498535, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7587496660432808, | |
| "grad_norm": 13.674204223054398, | |
| "learning_rate": 8.271734841028552e-08, | |
| "logits/chosen": -4.590233325958252, | |
| "logits/rejected": -4.405337333679199, | |
| "logps/chosen": -1.3775599002838135, | |
| "logps/rejected": -1.3799619674682617, | |
| "loss": 1.6717, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -3.4438998699188232, | |
| "rewards/margins": 0.0060052452608942986, | |
| "rewards/rejected": -3.4499053955078125, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.7694362810579749, | |
| "grad_norm": 8.839622500870057, | |
| "learning_rate": 7.588756023130833e-08, | |
| "logits/chosen": -4.494322776794434, | |
| "logits/rejected": -4.6520586013793945, | |
| "logps/chosen": -1.43705153465271, | |
| "logps/rejected": -1.5561714172363281, | |
| "loss": 1.6713, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -3.5926284790039062, | |
| "rewards/margins": 0.2977997660636902, | |
| "rewards/rejected": -3.890428066253662, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7801228960726689, | |
| "grad_norm": 7.5612671070202895, | |
| "learning_rate": 6.930128404315214e-08, | |
| "logits/chosen": -4.652656555175781, | |
| "logits/rejected": -4.5095977783203125, | |
| "logps/chosen": -1.4330899715423584, | |
| "logps/rejected": -1.393972635269165, | |
| "loss": 1.6884, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -3.5827248096466064, | |
| "rewards/margins": -0.09779311716556549, | |
| "rewards/rejected": -3.484931468963623, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.7908095110873631, | |
| "grad_norm": 10.689889310789306, | |
| "learning_rate": 6.296773134861824e-08, | |
| "logits/chosen": -4.595254421234131, | |
| "logits/rejected": -4.460488796234131, | |
| "logps/chosen": -1.455096960067749, | |
| "logps/rejected": -1.4411834478378296, | |
| "loss": 1.7402, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.637742280960083, | |
| "rewards/margins": -0.034783538430929184, | |
| "rewards/rejected": -3.602959156036377, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8014961261020572, | |
| "grad_norm": 11.234917641139386, | |
| "learning_rate": 5.6895760193850145e-08, | |
| "logits/chosen": -4.608285903930664, | |
| "logits/rejected": -4.652877330780029, | |
| "logps/chosen": -1.3920633792877197, | |
| "logps/rejected": -1.4802567958831787, | |
| "loss": 1.6655, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -3.480158567428589, | |
| "rewards/margins": 0.22048339247703552, | |
| "rewards/rejected": -3.7006423473358154, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 9.88912726576546, | |
| "learning_rate": 5.109386277955477e-08, | |
| "logits/chosen": -4.704692840576172, | |
| "logits/rejected": -4.580458641052246, | |
| "logps/chosen": -1.4294003248214722, | |
| "logps/rejected": -1.4911302328109741, | |
| "loss": 1.6673, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.573500871658325, | |
| "rewards/margins": 0.15432511270046234, | |
| "rewards/rejected": -3.72782564163208, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8228693561314454, | |
| "grad_norm": 10.411305158642252, | |
| "learning_rate": 4.557015358389216e-08, | |
| "logits/chosen": -4.629119396209717, | |
| "logits/rejected": -4.591732978820801, | |
| "logps/chosen": -1.371460199356079, | |
| "logps/rejected": -1.4805608987808228, | |
| "loss": 1.6999, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -3.4286506175994873, | |
| "rewards/margins": 0.27275168895721436, | |
| "rewards/rejected": -3.701402187347412, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.8335559711461394, | |
| "grad_norm": 10.400854396409033, | |
| "learning_rate": 4.0332358013644015e-08, | |
| "logits/chosen": -4.841821670532227, | |
| "logits/rejected": -4.738966941833496, | |
| "logps/chosen": -1.4363723993301392, | |
| "logps/rejected": -1.4949367046356201, | |
| "loss": 1.6777, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -3.590930938720703, | |
| "rewards/margins": 0.14641080796718597, | |
| "rewards/rejected": -3.7373416423797607, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8442425861608336, | |
| "grad_norm": 7.772358738977712, | |
| "learning_rate": 3.538780159953347e-08, | |
| "logits/chosen": -4.330094337463379, | |
| "logits/rejected": -4.12509298324585, | |
| "logps/chosen": -1.3891313076019287, | |
| "logps/rejected": -1.3778597116470337, | |
| "loss": 1.7104, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -3.4728283882141113, | |
| "rewards/margins": -0.02817906066775322, | |
| "rewards/rejected": -3.4446492195129395, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.8549292011755276, | |
| "grad_norm": 8.138330588799487, | |
| "learning_rate": 3.074339975080836e-08, | |
| "logits/chosen": -4.4906816482543945, | |
| "logits/rejected": -4.360891819000244, | |
| "logps/chosen": -1.4243663549423218, | |
| "logps/rejected": -1.5000258684158325, | |
| "loss": 1.6693, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -3.5609164237976074, | |
| "rewards/margins": 0.18914875388145447, | |
| "rewards/rejected": -3.7500648498535156, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8549292011755276, | |
| "eval_logits/chosen": -5.357321739196777, | |
| "eval_logits/rejected": -5.259011745452881, | |
| "eval_logps/chosen": -1.3368273973464966, | |
| "eval_logps/rejected": -1.3893804550170898, | |
| "eval_loss": 1.6597751379013062, | |
| "eval_rewards/accuracies": 0.5282257795333862, | |
| "eval_rewards/chosen": -3.342068672180176, | |
| "eval_rewards/margins": 0.13138233125209808, | |
| "eval_rewards/rejected": -3.4734511375427246, | |
| "eval_runtime": 31.809, | |
| "eval_samples_per_second": 61.649, | |
| "eval_steps_per_second": 1.949, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8656158161902218, | |
| "grad_norm": 9.025721024350105, | |
| "learning_rate": 2.6405648083415833e-08, | |
| "logits/chosen": -4.550287246704102, | |
| "logits/rejected": -4.305168628692627, | |
| "logps/chosen": -1.3779646158218384, | |
| "logps/rejected": -1.3646600246429443, | |
| "loss": 1.7228, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -3.444911479949951, | |
| "rewards/margins": -0.033261336386203766, | |
| "rewards/rejected": -3.4116501808166504, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.8763024312049158, | |
| "grad_norm": 11.89923381726874, | |
| "learning_rate": 2.2380613335296033e-08, | |
| "logits/chosen": -4.46389102935791, | |
| "logits/rejected": -4.445181846618652, | |
| "logps/chosen": -1.3117091655731201, | |
| "logps/rejected": -1.3666512966156006, | |
| "loss": 1.6743, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.2792727947235107, | |
| "rewards/margins": 0.13735508918762207, | |
| "rewards/rejected": -3.416628360748291, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.88698904621961, | |
| "grad_norm": 9.427141264972793, | |
| "learning_rate": 1.8673924881500823e-08, | |
| "logits/chosen": -4.327395439147949, | |
| "logits/rejected": -4.3094940185546875, | |
| "logps/chosen": -1.3074382543563843, | |
| "logps/rejected": -1.3440717458724976, | |
| "loss": 1.6969, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -3.2685959339141846, | |
| "rewards/margins": 0.09158400446176529, | |
| "rewards/rejected": -3.3601794242858887, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.897675661234304, | |
| "grad_norm": 8.90218297589139, | |
| "learning_rate": 1.5290766861003475e-08, | |
| "logits/chosen": -4.604608058929443, | |
| "logits/rejected": -4.5904130935668945, | |
| "logps/chosen": -1.3577730655670166, | |
| "logps/rejected": -1.4380416870117188, | |
| "loss": 1.6707, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -3.394432544708252, | |
| "rewards/margins": 0.20067182183265686, | |
| "rewards/rejected": -3.595104694366455, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9083622762489981, | |
| "grad_norm": 10.88127447829711, | |
| "learning_rate": 1.2235870926211616e-08, | |
| "logits/chosen": -4.505583763122559, | |
| "logits/rejected": -4.377171516418457, | |
| "logps/chosen": -1.379010558128357, | |
| "logps/rejected": -1.4630334377288818, | |
| "loss": 1.6586, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -3.447526454925537, | |
| "rewards/margins": 0.21005713939666748, | |
| "rewards/rejected": -3.6575839519500732, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.9190488912636923, | |
| "grad_norm": 10.139040628163542, | |
| "learning_rate": 9.513509625323518e-09, | |
| "logits/chosen": -4.474516868591309, | |
| "logits/rejected": -4.397044658660889, | |
| "logps/chosen": -1.3109729290008545, | |
| "logps/rejected": -1.3905436992645264, | |
| "loss": 1.6431, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -3.2774322032928467, | |
| "rewards/margins": 0.19892734289169312, | |
| "rewards/rejected": -3.4763596057891846, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9297355062783863, | |
| "grad_norm": 9.603243560454548, | |
| "learning_rate": 7.127490426783123e-09, | |
| "logits/chosen": -4.4349188804626465, | |
| "logits/rejected": -4.401902675628662, | |
| "logps/chosen": -1.382652997970581, | |
| "logps/rejected": -1.4352567195892334, | |
| "loss": 1.6374, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -3.456632137298584, | |
| "rewards/margins": 0.13150940835475922, | |
| "rewards/rejected": -3.588141918182373, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.9404221212930804, | |
| "grad_norm": 9.022313878003665, | |
| "learning_rate": 5.08115039419113e-09, | |
| "logits/chosen": -4.515431880950928, | |
| "logits/rejected": -4.496222496032715, | |
| "logps/chosen": -1.3486416339874268, | |
| "logps/rejected": -1.328399896621704, | |
| "loss": 1.6747, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -3.3716042041778564, | |
| "rewards/margins": -0.05060465261340141, | |
| "rewards/rejected": -3.3209996223449707, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9511087363077745, | |
| "grad_norm": 10.793631217244918, | |
| "learning_rate": 3.3773515191196646e-09, | |
| "logits/chosen": -4.495880126953125, | |
| "logits/rejected": -4.391939640045166, | |
| "logps/chosen": -1.355960726737976, | |
| "logps/rejected": -1.3439215421676636, | |
| "loss": 1.6919, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.389902114868164, | |
| "rewards/margins": -0.03009791299700737, | |
| "rewards/rejected": -3.3598036766052246, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.9617953513224686, | |
| "grad_norm": 11.320376659575638, | |
| "learning_rate": 2.0184767183584474e-09, | |
| "logits/chosen": -4.227232456207275, | |
| "logits/rejected": -4.283637523651123, | |
| "logps/chosen": -1.4779750108718872, | |
| "logps/rejected": -1.5757644176483154, | |
| "loss": 1.6992, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -3.694937229156494, | |
| "rewards/margins": 0.24447353184223175, | |
| "rewards/rejected": -3.939411163330078, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9724819663371627, | |
| "grad_norm": 9.391067862633497, | |
| "learning_rate": 1.0064265011902328e-09, | |
| "logits/chosen": -4.655524253845215, | |
| "logits/rejected": -4.6156415939331055, | |
| "logps/chosen": -1.4518479108810425, | |
| "logps/rejected": -1.3639124631881714, | |
| "loss": 1.7101, | |
| "rewards/accuracies": 0.3125, | |
| "rewards/chosen": -3.62962007522583, | |
| "rewards/margins": -0.21983873844146729, | |
| "rewards/rejected": -3.4097812175750732, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.9831685813518568, | |
| "grad_norm": 9.291539758676725, | |
| "learning_rate": 3.4261631135654167e-10, | |
| "logits/chosen": -4.43479585647583, | |
| "logits/rejected": -4.606427192687988, | |
| "logps/chosen": -1.3911080360412598, | |
| "logps/rejected": -1.5079371929168701, | |
| "loss": 1.6583, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -3.4777698516845703, | |
| "rewards/margins": 0.29207324981689453, | |
| "rewards/rejected": -3.769843339920044, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.9938551963665508, | |
| "grad_norm": 9.789963041329289, | |
| "learning_rate": 2.797454743164174e-11, | |
| "logits/chosen": -4.726589679718018, | |
| "logits/rejected": -4.504673957824707, | |
| "logps/chosen": -1.3526123762130737, | |
| "logps/rejected": -1.3742250204086304, | |
| "loss": 1.6691, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -3.38153076171875, | |
| "rewards/margins": 0.05403166264295578, | |
| "rewards/rejected": -3.4355628490448, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.9981298423724285, | |
| "step": 467, | |
| "total_flos": 0.0, | |
| "train_loss": 1.7099600828689476, | |
| "train_runtime": 3104.7014, | |
| "train_samples_per_second": 19.286, | |
| "train_steps_per_second": 0.15 | |
| } | |
| ], | |
| "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 | |
| } | |