| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 1875, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 2.6595744680851065e-08, | |
| "logits/chosen": -0.1435682475566864, | |
| "logits/rejected": -0.2822338938713074, | |
| "logps/chosen": -552.8109130859375, | |
| "logps/rejected": -630.5926513671875, | |
| "loss": 0.247, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.6595744680851066e-07, | |
| "logits/chosen": -0.16537699103355408, | |
| "logits/rejected": -0.1148439273238182, | |
| "logps/chosen": -533.032470703125, | |
| "logps/rejected": -578.028564453125, | |
| "loss": 0.2072, | |
| "rewards/accuracies": 0.375, | |
| "rewards/chosen": 8.621174492873251e-05, | |
| "rewards/margins": 2.2910655388841406e-05, | |
| "rewards/rejected": 6.330109317786992e-05, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5.319148936170213e-07, | |
| "logits/chosen": -0.17788179218769073, | |
| "logits/rejected": -0.07320378720760345, | |
| "logps/chosen": -506.20989990234375, | |
| "logps/rejected": -561.3789672851562, | |
| "loss": 0.2124, | |
| "rewards/accuracies": 0.38749998807907104, | |
| "rewards/chosen": -0.00013401822070591152, | |
| "rewards/margins": 1.3708593542105518e-05, | |
| "rewards/rejected": -0.00014772682334296405, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 7.97872340425532e-07, | |
| "logits/chosen": -0.19918851554393768, | |
| "logits/rejected": -0.13439549505710602, | |
| "logps/chosen": -486.73468017578125, | |
| "logps/rejected": -483.3734436035156, | |
| "loss": 0.2112, | |
| "rewards/accuracies": 0.4000000059604645, | |
| "rewards/chosen": -0.0002551859070081264, | |
| "rewards/margins": -5.1160714065190405e-05, | |
| "rewards/rejected": -0.00020402518566697836, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.0638297872340427e-06, | |
| "logits/chosen": -0.12800553441047668, | |
| "logits/rejected": -0.045729465782642365, | |
| "logps/chosen": -474.02825927734375, | |
| "logps/rejected": -471.7972717285156, | |
| "loss": 0.1915, | |
| "rewards/accuracies": 0.39375001192092896, | |
| "rewards/chosen": -0.0005546818720176816, | |
| "rewards/margins": -1.8026912584900856e-05, | |
| "rewards/rejected": -0.0005366549594327807, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.3297872340425533e-06, | |
| "logits/chosen": -0.16574732959270477, | |
| "logits/rejected": -0.09961490333080292, | |
| "logps/chosen": -512.2724609375, | |
| "logps/rejected": -547.2150268554688, | |
| "loss": 0.2085, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.00024178791500162333, | |
| "rewards/margins": 0.00024111787206493318, | |
| "rewards/rejected": -0.00048290580161847174, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.595744680851064e-06, | |
| "logits/chosen": -0.10868887603282928, | |
| "logits/rejected": -0.14005155861377716, | |
| "logps/chosen": -513.1686401367188, | |
| "logps/rejected": -552.2679443359375, | |
| "loss": 0.1949, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.00046477210707962513, | |
| "rewards/margins": 0.00030607733060605824, | |
| "rewards/rejected": -0.0007708495249971747, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.8617021276595745e-06, | |
| "logits/chosen": -0.1557120382785797, | |
| "logits/rejected": -0.14936389029026031, | |
| "logps/chosen": -495.12396240234375, | |
| "logps/rejected": -488.8611755371094, | |
| "loss": 0.2186, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -0.0014903920236974955, | |
| "rewards/margins": 6.922415923327208e-05, | |
| "rewards/rejected": -0.0015596161829307675, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.1276595744680853e-06, | |
| "logits/chosen": -0.09320048987865448, | |
| "logits/rejected": -0.19148024916648865, | |
| "logps/chosen": -500.7992248535156, | |
| "logps/rejected": -555.9862060546875, | |
| "loss": 0.2, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -0.0017530046170577407, | |
| "rewards/margins": 0.0014083541464060545, | |
| "rewards/rejected": -0.003161358879879117, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.393617021276596e-06, | |
| "logits/chosen": -0.1890203058719635, | |
| "logits/rejected": -0.10575833171606064, | |
| "logps/chosen": -448.5530700683594, | |
| "logps/rejected": -470.9744567871094, | |
| "loss": 0.209, | |
| "rewards/accuracies": 0.40625, | |
| "rewards/chosen": -0.003437761915847659, | |
| "rewards/margins": 0.0008986371685750782, | |
| "rewards/rejected": -0.0043363990262150764, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.6595744680851065e-06, | |
| "logits/chosen": -0.1906287968158722, | |
| "logits/rejected": -0.15528397262096405, | |
| "logps/chosen": -502.1371154785156, | |
| "logps/rejected": -580.352294921875, | |
| "loss": 0.1987, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.005366797558963299, | |
| "rewards/margins": 0.003352800849825144, | |
| "rewards/rejected": -0.00871959887444973, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9255319148936174e-06, | |
| "logits/chosen": -0.1939711570739746, | |
| "logits/rejected": -0.13397586345672607, | |
| "logps/chosen": -510.639404296875, | |
| "logps/rejected": -555.0497436523438, | |
| "loss": 0.2059, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.01002412661910057, | |
| "rewards/margins": 0.003309743944555521, | |
| "rewards/rejected": -0.013333870097994804, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.191489361702128e-06, | |
| "logits/chosen": -0.13544224202632904, | |
| "logits/rejected": -0.1556098461151123, | |
| "logps/chosen": -558.7181396484375, | |
| "logps/rejected": -572.3422241210938, | |
| "loss": 0.1984, | |
| "rewards/accuracies": 0.4000000059604645, | |
| "rewards/chosen": -0.016881179064512253, | |
| "rewards/margins": 0.001452689990401268, | |
| "rewards/rejected": -0.01833386905491352, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.457446808510639e-06, | |
| "logits/chosen": -0.23009908199310303, | |
| "logits/rejected": -0.06849310547113419, | |
| "logps/chosen": -488.73760986328125, | |
| "logps/rejected": -536.1317138671875, | |
| "loss": 0.2037, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.013895799405872822, | |
| "rewards/margins": 0.0098473085090518, | |
| "rewards/rejected": -0.02374310791492462, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.723404255319149e-06, | |
| "logits/chosen": -0.22787579894065857, | |
| "logits/rejected": -0.22026856243610382, | |
| "logps/chosen": -506.5633850097656, | |
| "logps/rejected": -527.6223754882812, | |
| "loss": 0.2149, | |
| "rewards/accuracies": 0.41874998807907104, | |
| "rewards/chosen": -0.023789843544363976, | |
| "rewards/margins": 0.007460168097168207, | |
| "rewards/rejected": -0.031250011175870895, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3.98936170212766e-06, | |
| "logits/chosen": -0.19232454895973206, | |
| "logits/rejected": -0.20156431198120117, | |
| "logps/chosen": -561.6283569335938, | |
| "logps/rejected": -535.9696655273438, | |
| "loss": 0.2204, | |
| "rewards/accuracies": 0.40625, | |
| "rewards/chosen": -0.03427774831652641, | |
| "rewards/margins": 0.0032845172099769115, | |
| "rewards/rejected": -0.03756226599216461, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.255319148936171e-06, | |
| "logits/chosen": -0.12956355512142181, | |
| "logits/rejected": -0.2759852707386017, | |
| "logps/chosen": -593.6724243164062, | |
| "logps/rejected": -623.8087158203125, | |
| "loss": 0.204, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.03722642734646797, | |
| "rewards/margins": 0.012259968556463718, | |
| "rewards/rejected": -0.04948639124631882, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.521276595744681e-06, | |
| "logits/chosen": -0.29936036467552185, | |
| "logits/rejected": -0.23110389709472656, | |
| "logps/chosen": -537.4310913085938, | |
| "logps/rejected": -559.8766479492188, | |
| "loss": 0.2115, | |
| "rewards/accuracies": 0.4312500059604645, | |
| "rewards/chosen": -0.04294046759605408, | |
| "rewards/margins": 0.013050531037151814, | |
| "rewards/rejected": -0.05599099397659302, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.787234042553192e-06, | |
| "logits/chosen": -0.22940175235271454, | |
| "logits/rejected": -0.15090861916542053, | |
| "logps/chosen": -498.94219970703125, | |
| "logps/rejected": -544.394287109375, | |
| "loss": 0.1869, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.03456363454461098, | |
| "rewards/margins": 0.021605905145406723, | |
| "rewards/rejected": -0.0561695322394371, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.999982660399688e-06, | |
| "logits/chosen": -0.25657087564468384, | |
| "logits/rejected": -0.2293904572725296, | |
| "logps/chosen": -565.2684936523438, | |
| "logps/rejected": -628.2935791015625, | |
| "loss": 0.1942, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.05312346667051315, | |
| "rewards/margins": 0.03432563692331314, | |
| "rewards/rejected": -0.0874490961432457, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.99937579964398e-06, | |
| "logits/chosen": -0.27247872948646545, | |
| "logits/rejected": -0.16829442977905273, | |
| "logps/chosen": -588.1021728515625, | |
| "logps/rejected": -624.6795654296875, | |
| "loss": 0.2013, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.07817868888378143, | |
| "rewards/margins": 0.012142921797931194, | |
| "rewards/rejected": -0.0903216004371643, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.9979021993870645e-06, | |
| "logits/chosen": -0.1175961121916771, | |
| "logits/rejected": -0.19183820486068726, | |
| "logps/chosen": -587.1693115234375, | |
| "logps/rejected": -589.8121337890625, | |
| "loss": 0.1982, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.05276166647672653, | |
| "rewards/margins": 0.009214815683662891, | |
| "rewards/rejected": -0.06197648122906685, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.995562370647553e-06, | |
| "logits/chosen": -0.1543715000152588, | |
| "logits/rejected": -0.12788452208042145, | |
| "logps/chosen": -565.2959594726562, | |
| "logps/rejected": -637.3449096679688, | |
| "loss": 0.2083, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.036317385733127594, | |
| "rewards/margins": 0.019043080508708954, | |
| "rewards/rejected": -0.05536046624183655, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.992357124836838e-06, | |
| "logits/chosen": -0.24705199897289276, | |
| "logits/rejected": -0.26284027099609375, | |
| "logps/chosen": -516.439453125, | |
| "logps/rejected": -574.214111328125, | |
| "loss": 0.1917, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.028448155149817467, | |
| "rewards/margins": 0.02016386389732361, | |
| "rewards/rejected": -0.048612017184495926, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.9882875734777044e-06, | |
| "logits/chosen": -0.2014235258102417, | |
| "logits/rejected": -0.05569050833582878, | |
| "logps/chosen": -596.5665893554688, | |
| "logps/rejected": -604.5208129882812, | |
| "loss": 0.2009, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.032715924084186554, | |
| "rewards/margins": 0.013567534275352955, | |
| "rewards/rejected": -0.046283453702926636, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.983355127818882e-06, | |
| "logits/chosen": -0.15398232638835907, | |
| "logits/rejected": -0.1951122283935547, | |
| "logps/chosen": -564.0726318359375, | |
| "logps/rejected": -575.0238037109375, | |
| "loss": 0.2, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -0.045795269310474396, | |
| "rewards/margins": 0.016760127618908882, | |
| "rewards/rejected": -0.06255538761615753, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.977561498345639e-06, | |
| "logits/chosen": -0.22048966586589813, | |
| "logits/rejected": -0.14720526337623596, | |
| "logps/chosen": -541.268310546875, | |
| "logps/rejected": -575.94482421875, | |
| "loss": 0.1918, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.039269499480724335, | |
| "rewards/margins": 0.02872983179986477, | |
| "rewards/rejected": -0.06799933314323425, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.970908694186624e-06, | |
| "logits/chosen": -0.12767961621284485, | |
| "logits/rejected": -0.17970682680606842, | |
| "logps/chosen": -586.6306762695312, | |
| "logps/rejected": -654.3707275390625, | |
| "loss": 0.1793, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.050308506935834885, | |
| "rewards/margins": 0.04710409790277481, | |
| "rewards/rejected": -0.0974126011133194, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.9633990224171305e-06, | |
| "logits/chosen": -0.246888667345047, | |
| "logits/rejected": -0.23898451030254364, | |
| "logps/chosen": -540.099609375, | |
| "logps/rejected": -619.6549072265625, | |
| "loss": 0.1998, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -0.05168341472744942, | |
| "rewards/margins": 0.023662500083446503, | |
| "rewards/rejected": -0.07534591108560562, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.955035087259046e-06, | |
| "logits/chosen": -0.17880979180335999, | |
| "logits/rejected": -0.17022277414798737, | |
| "logps/chosen": -519.5186767578125, | |
| "logps/rejected": -577.2052001953125, | |
| "loss": 0.1936, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.036568932235240936, | |
| "rewards/margins": 0.0271470807492733, | |
| "rewards/rejected": -0.06371601670980453, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.945819789177756e-06, | |
| "logits/chosen": -0.22478008270263672, | |
| "logits/rejected": -0.12748220562934875, | |
| "logps/chosen": -538.5833129882812, | |
| "logps/rejected": -555.4600830078125, | |
| "loss": 0.2204, | |
| "rewards/accuracies": 0.41874998807907104, | |
| "rewards/chosen": -0.0454036220908165, | |
| "rewards/margins": 0.00936979427933693, | |
| "rewards/rejected": -0.05477341264486313, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.935756323876306e-06, | |
| "logits/chosen": -0.16740605235099792, | |
| "logits/rejected": -0.1558527946472168, | |
| "logps/chosen": -514.8778076171875, | |
| "logps/rejected": -554.3466796875, | |
| "loss": 0.2064, | |
| "rewards/accuracies": 0.4437499940395355, | |
| "rewards/chosen": -0.03959917649626732, | |
| "rewards/margins": 0.016423584893345833, | |
| "rewards/rejected": -0.056022755801677704, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.924848181187199e-06, | |
| "logits/chosen": -0.21729227900505066, | |
| "logits/rejected": -0.13360168039798737, | |
| "logps/chosen": -550.1074829101562, | |
| "logps/rejected": -548.8150634765625, | |
| "loss": 0.2088, | |
| "rewards/accuracies": 0.41874998807907104, | |
| "rewards/chosen": -0.03942955285310745, | |
| "rewards/margins": 0.015313336625695229, | |
| "rewards/rejected": -0.05474288389086723, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.913099143862173e-06, | |
| "logits/chosen": -0.1903410106897354, | |
| "logits/rejected": -0.047067493200302124, | |
| "logps/chosen": -524.9461669921875, | |
| "logps/rejected": -586.0892333984375, | |
| "loss": 0.1892, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.039661504328250885, | |
| "rewards/margins": 0.02599067986011505, | |
| "rewards/rejected": -0.06565217673778534, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.900513286260416e-06, | |
| "logits/chosen": -0.13064715266227722, | |
| "logits/rejected": -0.11442460119724274, | |
| "logps/chosen": -598.4253540039062, | |
| "logps/rejected": -655.3058471679688, | |
| "loss": 0.2005, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.06107966974377632, | |
| "rewards/margins": 0.02944565750658512, | |
| "rewards/rejected": -0.09052532911300659, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.887094972935645e-06, | |
| "logits/chosen": -0.19566695392131805, | |
| "logits/rejected": -0.12640057504177094, | |
| "logps/chosen": -542.46533203125, | |
| "logps/rejected": -590.942626953125, | |
| "loss": 0.1869, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.046198152005672455, | |
| "rewards/margins": 0.034528639167547226, | |
| "rewards/rejected": -0.08072678744792938, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.87284885712256e-06, | |
| "logits/chosen": -0.15978778898715973, | |
| "logits/rejected": -0.08649591356515884, | |
| "logps/chosen": -535.330810546875, | |
| "logps/rejected": -571.8146362304688, | |
| "loss": 0.2087, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.05062000826001167, | |
| "rewards/margins": 0.016931967809796333, | |
| "rewards/rejected": -0.06755197793245316, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.857779879123181e-06, | |
| "logits/chosen": -0.14306055009365082, | |
| "logits/rejected": -0.18863961100578308, | |
| "logps/chosen": -550.8839111328125, | |
| "logps/rejected": -664.7866821289062, | |
| "loss": 0.1922, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.04616117477416992, | |
| "rewards/margins": 0.050246696919202805, | |
| "rewards/rejected": -0.09640787541866302, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.841893264593643e-06, | |
| "logits/chosen": -0.14398136734962463, | |
| "logits/rejected": -0.12858401238918304, | |
| "logps/chosen": -536.4591674804688, | |
| "logps/rejected": -554.4386596679688, | |
| "loss": 0.1966, | |
| "rewards/accuracies": 0.4437499940395355, | |
| "rewards/chosen": -0.046029798686504364, | |
| "rewards/margins": 0.020984839648008347, | |
| "rewards/rejected": -0.0670146495103836, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.825194522732023e-06, | |
| "logits/chosen": -0.16872592270374298, | |
| "logits/rejected": -0.04804924130439758, | |
| "logps/chosen": -523.2348022460938, | |
| "logps/rejected": -565.3974609375, | |
| "loss": 0.1916, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.04487336426973343, | |
| "rewards/margins": 0.028753217309713364, | |
| "rewards/rejected": -0.07362658530473709, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.807689444367853e-06, | |
| "logits/chosen": -0.16394059360027313, | |
| "logits/rejected": -0.07825560867786407, | |
| "logps/chosen": -569.9908447265625, | |
| "logps/rejected": -620.1237182617188, | |
| "loss": 0.1977, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.05317618325352669, | |
| "rewards/margins": 0.03387502580881119, | |
| "rewards/rejected": -0.08705119788646698, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.78938409995396e-06, | |
| "logits/chosen": -0.15660491585731506, | |
| "logits/rejected": -0.05097333714365959, | |
| "logps/chosen": -575.72607421875, | |
| "logps/rejected": -647.863037109375, | |
| "loss": 0.1964, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.04157186672091484, | |
| "rewards/margins": 0.042187802493572235, | |
| "rewards/rejected": -0.08375968039035797, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.770284837461342e-06, | |
| "logits/chosen": -0.1467301845550537, | |
| "logits/rejected": -0.041234057396650314, | |
| "logps/chosen": -506.21075439453125, | |
| "logps/rejected": -573.1678466796875, | |
| "loss": 0.1813, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.03669622167944908, | |
| "rewards/margins": 0.04058472067117691, | |
| "rewards/rejected": -0.07728095352649689, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.7503982801778015e-06, | |
| "logits/chosen": -0.049983613193035126, | |
| "logits/rejected": -0.09682069718837738, | |
| "logps/chosen": -542.2359619140625, | |
| "logps/rejected": -583.8301391601562, | |
| "loss": 0.1926, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.04950016364455223, | |
| "rewards/margins": 0.027737829834222794, | |
| "rewards/rejected": -0.07723800092935562, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.729731324411104e-06, | |
| "logits/chosen": -0.170378640294075, | |
| "logits/rejected": -0.12245140224695206, | |
| "logps/chosen": -491.5496520996094, | |
| "logps/rejected": -573.3743896484375, | |
| "loss": 0.1994, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.0331207700073719, | |
| "rewards/margins": 0.03893864154815674, | |
| "rewards/rejected": -0.07205941528081894, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.7082911370974645e-06, | |
| "logits/chosen": -0.21018514037132263, | |
| "logits/rejected": -0.09919361770153046, | |
| "logps/chosen": -500.1617736816406, | |
| "logps/rejected": -528.3983154296875, | |
| "loss": 0.1829, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.03072548285126686, | |
| "rewards/margins": 0.03223457187414169, | |
| "rewards/rejected": -0.06296005100011826, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.68608515331618e-06, | |
| "logits/chosen": -0.103789784014225, | |
| "logits/rejected": -0.09898993372917175, | |
| "logps/chosen": -548.4000854492188, | |
| "logps/rejected": -562.94482421875, | |
| "loss": 0.2153, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.03969540819525719, | |
| "rewards/margins": 0.025323593989014626, | |
| "rewards/rejected": -0.06501900404691696, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.663121073711269e-06, | |
| "logits/chosen": -0.1721930205821991, | |
| "logits/rejected": -0.1523117572069168, | |
| "logps/chosen": -503.42657470703125, | |
| "logps/rejected": -561.2427368164062, | |
| "loss": 0.2054, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -0.0317218080163002, | |
| "rewards/margins": 0.039346493780612946, | |
| "rewards/rejected": -0.07106830924749374, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.63940686182103e-06, | |
| "logits/chosen": -0.10965099185705185, | |
| "logits/rejected": -0.08603724092245102, | |
| "logps/chosen": -558.77490234375, | |
| "logps/rejected": -615.6389770507812, | |
| "loss": 0.1821, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.03726140409708023, | |
| "rewards/margins": 0.03986964374780655, | |
| "rewards/rejected": -0.07713104784488678, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.614950741316425e-06, | |
| "logits/chosen": -0.16542614996433258, | |
| "logits/rejected": -0.024035543203353882, | |
| "logps/chosen": -553.4063720703125, | |
| "logps/rejected": -628.834716796875, | |
| "loss": 0.1962, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.033900871872901917, | |
| "rewards/margins": 0.026126693934202194, | |
| "rewards/rejected": -0.06002756953239441, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.589761193149254e-06, | |
| "logits/chosen": -0.13893479108810425, | |
| "logits/rejected": -0.03659617155790329, | |
| "logps/chosen": -565.3289794921875, | |
| "logps/rejected": -582.9500732421875, | |
| "loss": 0.2002, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.034871987998485565, | |
| "rewards/margins": 0.018284570425748825, | |
| "rewards/rejected": -0.05315656214952469, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.563846952611112e-06, | |
| "logits/chosen": -0.03639233857393265, | |
| "logits/rejected": -0.13991451263427734, | |
| "logps/chosen": -574.9199829101562, | |
| "logps/rejected": -610.3167114257812, | |
| "loss": 0.2025, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.028040558099746704, | |
| "rewards/margins": 0.02244061976671219, | |
| "rewards/rejected": -0.05048117786645889, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.537217006304141e-06, | |
| "logits/chosen": -0.1792822778224945, | |
| "logits/rejected": -0.08993463218212128, | |
| "logps/chosen": -525.2308349609375, | |
| "logps/rejected": -558.35791015625, | |
| "loss": 0.2095, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.022352229803800583, | |
| "rewards/margins": 0.015106442384421825, | |
| "rewards/rejected": -0.03745867311954498, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.50988058902464e-06, | |
| "logits/chosen": -0.10850954055786133, | |
| "logits/rejected": -0.0964195728302002, | |
| "logps/chosen": -534.2761840820312, | |
| "logps/rejected": -559.3724975585938, | |
| "loss": 0.2077, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.020403413102030754, | |
| "rewards/margins": 0.021993646398186684, | |
| "rewards/rejected": -0.04239705950021744, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.481847180560593e-06, | |
| "logits/chosen": -0.05597566440701485, | |
| "logits/rejected": -0.1112387552857399, | |
| "logps/chosen": -582.2784423828125, | |
| "logps/rejected": -618.896728515625, | |
| "loss": 0.197, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.032762907445430756, | |
| "rewards/margins": 0.018652746453881264, | |
| "rewards/rejected": -0.05141565203666687, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.453126502404253e-06, | |
| "logits/chosen": -0.06729540228843689, | |
| "logits/rejected": -0.045423537492752075, | |
| "logps/chosen": -526.5831298828125, | |
| "logps/rejected": -574.4923706054688, | |
| "loss": 0.1946, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.027647733688354492, | |
| "rewards/margins": 0.02961956337094307, | |
| "rewards/rejected": -0.05726729705929756, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.423728514380892e-06, | |
| "logits/chosen": -0.09008444845676422, | |
| "logits/rejected": 0.0276274923235178, | |
| "logps/chosen": -544.626708984375, | |
| "logps/rejected": -580.9148559570312, | |
| "loss": 0.2048, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.036577023565769196, | |
| "rewards/margins": 0.01596749946475029, | |
| "rewards/rejected": -0.052544523030519485, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.393663411194918e-06, | |
| "logits/chosen": -0.05868130922317505, | |
| "logits/rejected": -0.04863614961504936, | |
| "logps/chosen": -531.1651000976562, | |
| "logps/rejected": -553.3944702148438, | |
| "loss": 0.2092, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -0.03158984333276749, | |
| "rewards/margins": 0.02088005281984806, | |
| "rewards/rejected": -0.052469898015260696, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.362941618894523e-06, | |
| "logits/chosen": -0.08449320495128632, | |
| "logits/rejected": -0.012650961056351662, | |
| "logps/chosen": -541.2258911132812, | |
| "logps/rejected": -583.4708862304688, | |
| "loss": 0.2071, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.03278597444295883, | |
| "rewards/margins": 0.016275811940431595, | |
| "rewards/rejected": -0.04906178265810013, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.331573791256116e-06, | |
| "logits/chosen": -0.05021784454584122, | |
| "logits/rejected": -0.0996134951710701, | |
| "logps/chosen": -490.427978515625, | |
| "logps/rejected": -564.3570556640625, | |
| "loss": 0.1908, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.025160646066069603, | |
| "rewards/margins": 0.028088271617889404, | |
| "rewards/rejected": -0.053248919546604156, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.299570806089786e-06, | |
| "logits/chosen": 0.0037119381595402956, | |
| "logits/rejected": -0.03845081478357315, | |
| "logps/chosen": -517.3943481445312, | |
| "logps/rejected": -546.4298095703125, | |
| "loss": 0.1878, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.03542790561914444, | |
| "rewards/margins": 0.03082282282412052, | |
| "rewards/rejected": -0.06625071913003922, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.266943761467057e-06, | |
| "logits/chosen": -0.11214573681354523, | |
| "logits/rejected": -0.048879437148571014, | |
| "logps/chosen": -585.6790771484375, | |
| "logps/rejected": -597.4764404296875, | |
| "loss": 0.1957, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.04125026613473892, | |
| "rewards/margins": 0.022660288959741592, | |
| "rewards/rejected": -0.06391056627035141, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.233703971872287e-06, | |
| "logits/chosen": -0.03887901455163956, | |
| "logits/rejected": -0.041454676538705826, | |
| "logps/chosen": -554.0848999023438, | |
| "logps/rejected": -593.8275756835938, | |
| "loss": 0.1925, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.04528883099555969, | |
| "rewards/margins": 0.03674833104014397, | |
| "rewards/rejected": -0.08203716576099396, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.1998629642789925e-06, | |
| "logits/chosen": -0.14514538645744324, | |
| "logits/rejected": -0.12874510884284973, | |
| "logps/chosen": -580.5797729492188, | |
| "logps/rejected": -627.352294921875, | |
| "loss": 0.1972, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.053490232676267624, | |
| "rewards/margins": 0.03829261288046837, | |
| "rewards/rejected": -0.09178284555673599, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.165432474152505e-06, | |
| "logits/chosen": -0.10119237750768661, | |
| "logits/rejected": -0.12084090709686279, | |
| "logps/chosen": -556.23828125, | |
| "logps/rejected": -601.0460205078125, | |
| "loss": 0.1939, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.05898448824882507, | |
| "rewards/margins": 0.03952840715646744, | |
| "rewards/rejected": -0.09851288795471191, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.130424441380308e-06, | |
| "logits/chosen": -0.051937490701675415, | |
| "logits/rejected": -0.11804668605327606, | |
| "logps/chosen": -618.0116577148438, | |
| "logps/rejected": -681.9761962890625, | |
| "loss": 0.1944, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.08069413900375366, | |
| "rewards/margins": 0.053174007683992386, | |
| "rewards/rejected": -0.13386814296245575, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.09485100613151e-06, | |
| "logits/chosen": 0.03249294310808182, | |
| "logits/rejected": -0.03189364820718765, | |
| "logps/chosen": -587.2728881835938, | |
| "logps/rejected": -678.32568359375, | |
| "loss": 0.2106, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.06832827627658844, | |
| "rewards/margins": 0.0339791476726532, | |
| "rewards/rejected": -0.10230742394924164, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.058724504646834e-06, | |
| "logits/chosen": 0.003849644912406802, | |
| "logits/rejected": -0.07050515711307526, | |
| "logps/chosen": -485.66375732421875, | |
| "logps/rejected": -613.0921020507812, | |
| "loss": 0.1938, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.0477924570441246, | |
| "rewards/margins": 0.05937989428639412, | |
| "rewards/rejected": -0.10717234760522842, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.022057464960632e-06, | |
| "logits/chosen": -0.034047119319438934, | |
| "logits/rejected": 0.01398265827447176, | |
| "logps/chosen": -568.3038940429688, | |
| "logps/rejected": -603.4803466796875, | |
| "loss": 0.2027, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.04010879248380661, | |
| "rewards/margins": 0.03294765576720238, | |
| "rewards/rejected": -0.07305644452571869, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.984862602556383e-06, | |
| "logits/chosen": -0.0736241266131401, | |
| "logits/rejected": -0.03072969987988472, | |
| "logps/chosen": -584.6604614257812, | |
| "logps/rejected": -646.9960327148438, | |
| "loss": 0.1893, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.022528892382979393, | |
| "rewards/margins": 0.03268754482269287, | |
| "rewards/rejected": -0.055216431617736816, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.947152815957187e-06, | |
| "logits/chosen": -0.07144252955913544, | |
| "logits/rejected": -0.04082520306110382, | |
| "logps/chosen": -536.7813720703125, | |
| "logps/rejected": -604.7335205078125, | |
| "loss": 0.185, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.04123537242412567, | |
| "rewards/margins": 0.04173678532242775, | |
| "rewards/rejected": -0.08297216892242432, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.908941182252785e-06, | |
| "logits/chosen": -0.04457836225628853, | |
| "logits/rejected": -0.019069496542215347, | |
| "logps/chosen": -504.98260498046875, | |
| "logps/rejected": -545.35107421875, | |
| "loss": 0.201, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.03742678835988045, | |
| "rewards/margins": 0.03181907907128334, | |
| "rewards/rejected": -0.06924585998058319, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.8702409525646535e-06, | |
| "logits/chosen": -0.08643130213022232, | |
| "logits/rejected": -0.00629496481269598, | |
| "logps/chosen": -506.75128173828125, | |
| "logps/rejected": -579.1588745117188, | |
| "loss": 0.2058, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.03903960436582565, | |
| "rewards/margins": 0.03844233229756355, | |
| "rewards/rejected": -0.0774819403886795, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.8310655474507495e-06, | |
| "logits/chosen": -0.10445193201303482, | |
| "logits/rejected": -0.05900994688272476, | |
| "logps/chosen": -561.2862548828125, | |
| "logps/rejected": -646.6036987304688, | |
| "loss": 0.1828, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.03991767764091492, | |
| "rewards/margins": 0.04778400436043739, | |
| "rewards/rejected": -0.08770167827606201, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.7914285522515002e-06, | |
| "logits/chosen": -0.10235663503408432, | |
| "logits/rejected": -0.06297244131565094, | |
| "logps/chosen": -533.1697387695312, | |
| "logps/rejected": -621.258056640625, | |
| "loss": 0.1897, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.035120584070682526, | |
| "rewards/margins": 0.046875499188899994, | |
| "rewards/rejected": -0.08199608325958252, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.751343712378639e-06, | |
| "logits/chosen": 0.04809337109327316, | |
| "logits/rejected": -0.11011433601379395, | |
| "logps/chosen": -501.5302734375, | |
| "logps/rejected": -548.520263671875, | |
| "loss": 0.198, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.040851037949323654, | |
| "rewards/margins": 0.02717413939535618, | |
| "rewards/rejected": -0.06802517920732498, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.710824928548546e-06, | |
| "logits/chosen": -0.01540103554725647, | |
| "logits/rejected": 0.029544491320848465, | |
| "logps/chosen": -581.104736328125, | |
| "logps/rejected": -622.2809448242188, | |
| "loss": 0.1955, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.03365876525640488, | |
| "rewards/margins": 0.01963328942656517, | |
| "rewards/rejected": -0.05329205468297005, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.6698862519617225e-06, | |
| "logits/chosen": 0.05280674248933792, | |
| "logits/rejected": -0.0824475884437561, | |
| "logps/chosen": -551.7106323242188, | |
| "logps/rejected": -647.3377685546875, | |
| "loss": 0.1755, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.03773096203804016, | |
| "rewards/margins": 0.051321662962436676, | |
| "rewards/rejected": -0.08905263245105743, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.6285418794300793e-06, | |
| "logits/chosen": 0.035738877952098846, | |
| "logits/rejected": -0.04512631893157959, | |
| "logps/chosen": -562.4360961914062, | |
| "logps/rejected": -655.2886962890625, | |
| "loss": 0.1919, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.041907649487257004, | |
| "rewards/margins": 0.0431944876909256, | |
| "rewards/rejected": -0.0851021409034729, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.5868061484537365e-06, | |
| "logits/chosen": -0.029743824154138565, | |
| "logits/rejected": 0.053073786199092865, | |
| "logps/chosen": -578.3715209960938, | |
| "logps/rejected": -652.7194213867188, | |
| "loss": 0.1883, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.043207161128520966, | |
| "rewards/margins": 0.047172196209430695, | |
| "rewards/rejected": -0.09037934988737106, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.5446935322490285e-06, | |
| "logits/chosen": -0.006876747123897076, | |
| "logits/rejected": 0.015042210929095745, | |
| "logps/chosen": -525.9793090820312, | |
| "logps/rejected": -600.8638916015625, | |
| "loss": 0.1998, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.03883466124534607, | |
| "rewards/margins": 0.05057210475206375, | |
| "rewards/rejected": -0.08940677344799042, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.502218634729447e-06, | |
| "logits/chosen": -0.06274811178445816, | |
| "logits/rejected": 0.020462125539779663, | |
| "logps/chosen": -544.5744018554688, | |
| "logps/rejected": -636.985107421875, | |
| "loss": 0.199, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.02988346479833126, | |
| "rewards/margins": 0.046063631772994995, | |
| "rewards/rejected": -0.07594709098339081, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.459396185441265e-06, | |
| "logits/chosen": -0.034183721989393234, | |
| "logits/rejected": 0.007283804006874561, | |
| "logps/chosen": -590.5462646484375, | |
| "logps/rejected": -617.2237548828125, | |
| "loss": 0.1814, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.03479190170764923, | |
| "rewards/margins": 0.03565431758761406, | |
| "rewards/rejected": -0.07044622302055359, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.4162410344555834e-06, | |
| "logits/chosen": -0.031631968915462494, | |
| "logits/rejected": -0.03749103471636772, | |
| "logps/chosen": -553.4362182617188, | |
| "logps/rejected": -585.8868408203125, | |
| "loss": 0.1936, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.025158634409308434, | |
| "rewards/margins": 0.02824149653315544, | |
| "rewards/rejected": -0.053400129079818726, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.3727681472185937e-06, | |
| "logits/chosen": -0.017966533079743385, | |
| "logits/rejected": 0.07461138069629669, | |
| "logps/chosen": -504.60809326171875, | |
| "logps/rejected": -541.7296142578125, | |
| "loss": 0.2005, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.018680742010474205, | |
| "rewards/margins": 0.02315039001405239, | |
| "rewards/rejected": -0.0418311282992363, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.3289925993618217e-06, | |
| "logits/chosen": -0.05572665482759476, | |
| "logits/rejected": 0.0033084123861044645, | |
| "logps/chosen": -538.4072875976562, | |
| "logps/rejected": -589.1475219726562, | |
| "loss": 0.1734, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.017685720697045326, | |
| "rewards/margins": 0.02718517556786537, | |
| "rewards/rejected": -0.04487089440226555, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.2849295714741643e-06, | |
| "logits/chosen": -0.019068999215960503, | |
| "logits/rejected": -0.08031287044286728, | |
| "logps/chosen": -477.96832275390625, | |
| "logps/rejected": -538.3767700195312, | |
| "loss": 0.2003, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.017096657305955887, | |
| "rewards/margins": 0.022515540942549706, | |
| "rewards/rejected": -0.03961219638586044, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.2405943438375287e-06, | |
| "logits/chosen": -0.10866532474756241, | |
| "logits/rejected": -0.014922107569873333, | |
| "logps/chosen": -573.4918212890625, | |
| "logps/rejected": -601.1939086914062, | |
| "loss": 0.1981, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.030659830197691917, | |
| "rewards/margins": 0.022852713242173195, | |
| "rewards/rejected": -0.05351255089044571, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.1960022911279036e-06, | |
| "logits/chosen": -0.005284407641738653, | |
| "logits/rejected": -0.043508388102054596, | |
| "logps/chosen": -556.1735229492188, | |
| "logps/rejected": -627.1209106445312, | |
| "loss": 0.2072, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -0.0352107435464859, | |
| "rewards/margins": 0.038544684648513794, | |
| "rewards/rejected": -0.0737554281949997, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.1511688770836844e-06, | |
| "logits/chosen": -0.06900519132614136, | |
| "logits/rejected": -0.008347516879439354, | |
| "logps/chosen": -518.9674072265625, | |
| "logps/rejected": -563.190185546875, | |
| "loss": 0.1873, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.0382576659321785, | |
| "rewards/margins": 0.03923247009515762, | |
| "rewards/rejected": -0.07749013602733612, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.1061096491431307e-06, | |
| "logits/chosen": -0.004668307490646839, | |
| "logits/rejected": -0.05605720728635788, | |
| "logps/chosen": -560.2337646484375, | |
| "logps/rejected": -637.04931640625, | |
| "loss": 0.1747, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.06013607978820801, | |
| "rewards/margins": 0.052308034151792526, | |
| "rewards/rejected": -0.11244411766529083, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.0608402330527796e-06, | |
| "logits/chosen": -0.06335186958312988, | |
| "logits/rejected": -0.027245525270700455, | |
| "logps/chosen": -516.86865234375, | |
| "logps/rejected": -573.42724609375, | |
| "loss": 0.1955, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.06095459312200546, | |
| "rewards/margins": 0.03399496152997017, | |
| "rewards/rejected": -0.09494956582784653, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.0153763274487176e-06, | |
| "logits/chosen": -0.016411593183875084, | |
| "logits/rejected": 0.005086073186248541, | |
| "logps/chosen": -552.3294677734375, | |
| "logps/rejected": -607.4609375, | |
| "loss": 0.1793, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.06465917080640793, | |
| "rewards/margins": 0.038704242557287216, | |
| "rewards/rejected": -0.10336340963840485, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.9697336984125683e-06, | |
| "logits/chosen": -0.0311849657446146, | |
| "logits/rejected": -0.05022461339831352, | |
| "logps/chosen": -507.9979553222656, | |
| "logps/rejected": -572.2631225585938, | |
| "loss": 0.1788, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.03872525319457054, | |
| "rewards/margins": 0.03657948225736618, | |
| "rewards/rejected": -0.07530473172664642, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.923928174004094e-06, | |
| "logits/chosen": -0.1109534353017807, | |
| "logits/rejected": -0.057219941169023514, | |
| "logps/chosen": -550.4554443359375, | |
| "logps/rejected": -599.104736328125, | |
| "loss": 0.1928, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.05719093605875969, | |
| "rewards/margins": 0.02921503223478794, | |
| "rewards/rejected": -0.08640597015619278, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.8779756387723036e-06, | |
| "logits/chosen": -0.009409350343048573, | |
| "logits/rejected": 0.0022618831135332584, | |
| "logps/chosen": -604.61181640625, | |
| "logps/rejected": -644.1531982421875, | |
| "loss": 0.1886, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.05070986598730087, | |
| "rewards/margins": 0.03501267358660698, | |
| "rewards/rejected": -0.08572253584861755, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.831892028246968e-06, | |
| "logits/chosen": -0.06726431846618652, | |
| "logits/rejected": -0.08842380344867706, | |
| "logps/chosen": -606.5565185546875, | |
| "logps/rejected": -661.2351684570312, | |
| "loss": 0.2006, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.05755617097020149, | |
| "rewards/margins": 0.024521058425307274, | |
| "rewards/rejected": -0.08207723498344421, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.7856933234124617e-06, | |
| "logits/chosen": 0.021266190335154533, | |
| "logits/rejected": -0.03655446320772171, | |
| "logps/chosen": -579.4450073242188, | |
| "logps/rejected": -625.4588623046875, | |
| "loss": 0.1998, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.0589175745844841, | |
| "rewards/margins": 0.03393319249153137, | |
| "rewards/rejected": -0.09285076707601547, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.7393955451658387e-06, | |
| "logits/chosen": -0.001839430071413517, | |
| "logits/rejected": 0.02721802517771721, | |
| "logps/chosen": -569.8568115234375, | |
| "logps/rejected": -593.9893798828125, | |
| "loss": 0.1865, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.058799803256988525, | |
| "rewards/margins": 0.0315038226544857, | |
| "rewards/rejected": -0.09030362218618393, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.6930147487610667e-06, | |
| "logits/chosen": -0.014354798011481762, | |
| "logits/rejected": -0.015522453002631664, | |
| "logps/chosen": -546.56884765625, | |
| "logps/rejected": -624.5516967773438, | |
| "loss": 0.2, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.06721802055835724, | |
| "rewards/margins": 0.030059367418289185, | |
| "rewards/rejected": -0.09727738052606583, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.6465670182413487e-06, | |
| "logits/chosen": 0.009198218584060669, | |
| "logits/rejected": -0.031388141214847565, | |
| "logps/chosen": -546.1246337890625, | |
| "logps/rejected": -607.7100219726562, | |
| "loss": 0.1942, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.050740886479616165, | |
| "rewards/margins": 0.04155651852488518, | |
| "rewards/rejected": -0.09229740500450134, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.6000684608614594e-06, | |
| "logits/chosen": -0.1307494342327118, | |
| "logits/rejected": 0.043267279863357544, | |
| "logps/chosen": -562.8900146484375, | |
| "logps/rejected": -619.1513671875, | |
| "loss": 0.1872, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.06052188202738762, | |
| "rewards/margins": 0.034762654453516006, | |
| "rewards/rejected": -0.09528453648090363, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5535352015020338e-06, | |
| "logits/chosen": -0.09118489176034927, | |
| "logits/rejected": -0.04373834654688835, | |
| "logps/chosen": -546.9132690429688, | |
| "logps/rejected": -610.7948608398438, | |
| "loss": 0.1832, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.06443742662668228, | |
| "rewards/margins": 0.0450962632894516, | |
| "rewards/rejected": -0.10953368991613388, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.506983377077741e-06, | |
| "logits/chosen": -0.10389532893896103, | |
| "logits/rejected": 0.04294447600841522, | |
| "logps/chosen": -597.6129760742188, | |
| "logps/rejected": -630.375, | |
| "loss": 0.1955, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.050885915756225586, | |
| "rewards/margins": 0.04200722649693489, | |
| "rewards/rejected": -0.09289313852787018, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.460429130941289e-06, | |
| "logits/chosen": -0.05054370313882828, | |
| "logits/rejected": -0.041653625667095184, | |
| "logps/chosen": -562.7481689453125, | |
| "logps/rejected": -637.655029296875, | |
| "loss": 0.1907, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.06117384508252144, | |
| "rewards/margins": 0.04911119490861893, | |
| "rewards/rejected": -0.11028504371643066, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.413888607285192e-06, | |
| "logits/chosen": -0.13274100422859192, | |
| "logits/rejected": -0.03395792096853256, | |
| "logps/chosen": -592.2869262695312, | |
| "logps/rejected": -656.2505493164062, | |
| "loss": 0.1916, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.06654617935419083, | |
| "rewards/margins": 0.048628997057676315, | |
| "rewards/rejected": -0.11517517268657684, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.367377945543249e-06, | |
| "logits/chosen": -0.04836379736661911, | |
| "logits/rejected": 0.0366315022110939, | |
| "logps/chosen": -561.9847412109375, | |
| "logps/rejected": -609.58935546875, | |
| "loss": 0.1841, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.05483894422650337, | |
| "rewards/margins": 0.03531566262245178, | |
| "rewards/rejected": -0.09015461057424545, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.320913274793676e-06, | |
| "logits/chosen": -0.02548578381538391, | |
| "logits/rejected": -0.011551322415471077, | |
| "logps/chosen": -517.5578002929688, | |
| "logps/rejected": -587.1166381835938, | |
| "loss": 0.1911, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.03753050044178963, | |
| "rewards/margins": 0.05013161897659302, | |
| "rewards/rejected": -0.08766211569309235, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.27451070816582e-06, | |
| "logits/chosen": -0.07835321873426437, | |
| "logits/rejected": -0.008995351381599903, | |
| "logps/chosen": -603.7330932617188, | |
| "logps/rejected": -606.139404296875, | |
| "loss": 0.1808, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.05653967335820198, | |
| "rewards/margins": 0.02898581698536873, | |
| "rewards/rejected": -0.08552548289299011, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.228186337252414e-06, | |
| "logits/chosen": -0.05141784995794296, | |
| "logits/rejected": -0.023404929786920547, | |
| "logps/chosen": -539.2532958984375, | |
| "logps/rejected": -597.3132934570312, | |
| "loss": 0.1901, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.04464349150657654, | |
| "rewards/margins": 0.027696922421455383, | |
| "rewards/rejected": -0.07234041392803192, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.1819562265292946e-06, | |
| "logits/chosen": -0.024759747087955475, | |
| "logits/rejected": -0.13666841387748718, | |
| "logps/chosen": -521.7346801757812, | |
| "logps/rejected": -593.8477783203125, | |
| "loss": 0.1896, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.04558413475751877, | |
| "rewards/margins": 0.028859147801995277, | |
| "rewards/rejected": -0.0744432732462883, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.1358364077845236e-06, | |
| "logits/chosen": -0.1430703103542328, | |
| "logits/rejected": -0.04963989928364754, | |
| "logps/chosen": -554.0672607421875, | |
| "logps/rejected": -595.87939453125, | |
| "loss": 0.1941, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.04022881016135216, | |
| "rewards/margins": 0.03246025741100311, | |
| "rewards/rejected": -0.07268907129764557, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.089842874558849e-06, | |
| "logits/chosen": -0.010374871082603931, | |
| "logits/rejected": -0.07016284763813019, | |
| "logps/chosen": -510.901611328125, | |
| "logps/rejected": -578.7679443359375, | |
| "loss": 0.2011, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.03890245407819748, | |
| "rewards/margins": 0.032768718898296356, | |
| "rewards/rejected": -0.07167118042707443, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.0439915765994242e-06, | |
| "logits/chosen": -0.06316359341144562, | |
| "logits/rejected": 0.0023398040793836117, | |
| "logps/chosen": -512.5825805664062, | |
| "logps/rejected": -597.9288330078125, | |
| "loss": 0.1817, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.04139110818505287, | |
| "rewards/margins": 0.03969164192676544, | |
| "rewards/rejected": -0.08108274638652802, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.9982984143287186e-06, | |
| "logits/chosen": -0.0061457292176783085, | |
| "logits/rejected": -0.014603073708713055, | |
| "logps/chosen": -573.0762939453125, | |
| "logps/rejected": -628.219970703125, | |
| "loss": 0.1784, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.0488489493727684, | |
| "rewards/margins": 0.04788987711071968, | |
| "rewards/rejected": -0.09673883020877838, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.95277923333053e-06, | |
| "logits/chosen": -0.03414573520421982, | |
| "logits/rejected": -0.014127236790955067, | |
| "logps/chosen": -551.8556518554688, | |
| "logps/rejected": -587.9727783203125, | |
| "loss": 0.184, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.04027246683835983, | |
| "rewards/margins": 0.03260934352874756, | |
| "rewards/rejected": -0.07288181036710739, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.9074498188550156e-06, | |
| "logits/chosen": -0.040990687906742096, | |
| "logits/rejected": 0.027432560920715332, | |
| "logps/chosen": -519.9183349609375, | |
| "logps/rejected": -585.5509033203125, | |
| "loss": 0.1709, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.028263911604881287, | |
| "rewards/margins": 0.038620587438344955, | |
| "rewards/rejected": -0.06688449531793594, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.862325890344643e-06, | |
| "logits/chosen": -0.09657453745603561, | |
| "logits/rejected": -0.028930921107530594, | |
| "logps/chosen": -509.68731689453125, | |
| "logps/rejected": -545.7371215820312, | |
| "loss": 0.1907, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.03510227054357529, | |
| "rewards/margins": 0.028696129098534584, | |
| "rewards/rejected": -0.06379840523004532, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.817423095982972e-06, | |
| "logits/chosen": -0.001943819923326373, | |
| "logits/rejected": -0.10663602501153946, | |
| "logps/chosen": -541.8197021484375, | |
| "logps/rejected": -631.3278198242188, | |
| "loss": 0.183, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.05557183176279068, | |
| "rewards/margins": 0.04637579992413521, | |
| "rewards/rejected": -0.10194762796163559, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.7727570072681293e-06, | |
| "logits/chosen": -0.03551124408841133, | |
| "logits/rejected": -0.03228951245546341, | |
| "logps/chosen": -513.2832641601562, | |
| "logps/rejected": -647.9851684570312, | |
| "loss": 0.1855, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.03109642304480076, | |
| "rewards/margins": 0.06740632653236389, | |
| "rewards/rejected": -0.0985027551651001, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.7283431136128961e-06, | |
| "logits/chosen": -0.006384107284247875, | |
| "logits/rejected": -0.0014162927400320768, | |
| "logps/chosen": -545.4998168945312, | |
| "logps/rejected": -616.1009521484375, | |
| "loss": 0.1799, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.057301461696624756, | |
| "rewards/margins": 0.0508766695857048, | |
| "rewards/rejected": -0.10817812383174896, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.6841968169732478e-06, | |
| "logits/chosen": 0.031141120940446854, | |
| "logits/rejected": -0.0957186371088028, | |
| "logps/chosen": -535.7457275390625, | |
| "logps/rejected": -653.5030517578125, | |
| "loss": 0.1861, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.06120613217353821, | |
| "rewards/margins": 0.046815842390060425, | |
| "rewards/rejected": -0.10802197456359863, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.6403334265072284e-06, | |
| "logits/chosen": -0.06112520769238472, | |
| "logits/rejected": 0.02096955105662346, | |
| "logps/chosen": -497.9486389160156, | |
| "logps/rejected": -615.8877563476562, | |
| "loss": 0.1872, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.03988587483763695, | |
| "rewards/margins": 0.04784765467047691, | |
| "rewards/rejected": -0.08773352205753326, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.5967681532660066e-06, | |
| "logits/chosen": -0.07220903038978577, | |
| "logits/rejected": 0.042104389518499374, | |
| "logps/chosen": -510.8187561035156, | |
| "logps/rejected": -635.9632568359375, | |
| "loss": 0.1726, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.04127117991447449, | |
| "rewards/margins": 0.06924723088741302, | |
| "rewards/rejected": -0.11051841825246811, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.5535161049189463e-06, | |
| "logits/chosen": -0.14815333485603333, | |
| "logits/rejected": -0.03901112079620361, | |
| "logps/chosen": -537.5811157226562, | |
| "logps/rejected": -621.5980834960938, | |
| "loss": 0.1831, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.058847375214099884, | |
| "rewards/margins": 0.05733171105384827, | |
| "rewards/rejected": -0.11617908626794815, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.5105922805145356e-06, | |
| "logits/chosen": -0.029178712517023087, | |
| "logits/rejected": -0.11420316994190216, | |
| "logps/chosen": -573.003662109375, | |
| "logps/rejected": -637.9074096679688, | |
| "loss": 0.1971, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -0.07729945331811905, | |
| "rewards/margins": 0.04123348742723465, | |
| "rewards/rejected": -0.1185329332947731, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.4680115652789823e-06, | |
| "logits/chosen": -0.07775846868753433, | |
| "logits/rejected": -0.07050256431102753, | |
| "logps/chosen": -551.1271362304688, | |
| "logps/rejected": -654.5675659179688, | |
| "loss": 0.1784, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.06159217283129692, | |
| "rewards/margins": 0.07197807729244232, | |
| "rewards/rejected": -0.13357023894786835, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.4257887254542767e-06, | |
| "logits/chosen": -0.052437614649534225, | |
| "logits/rejected": -0.10926377773284912, | |
| "logps/chosen": -546.0693969726562, | |
| "logps/rejected": -625.9730834960938, | |
| "loss": 0.1806, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.053220171481370926, | |
| "rewards/margins": 0.04990285634994507, | |
| "rewards/rejected": -0.1031230241060257, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.3839384031775227e-06, | |
| "logits/chosen": 0.014421490021049976, | |
| "logits/rejected": -0.14018869400024414, | |
| "logps/chosen": -590.1350708007812, | |
| "logps/rejected": -659.4343872070312, | |
| "loss": 0.2018, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.051880426704883575, | |
| "rewards/margins": 0.05124548077583313, | |
| "rewards/rejected": -0.1031259074807167, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.342475111403298e-06, | |
| "logits/chosen": -0.0805274024605751, | |
| "logits/rejected": -0.01422274298965931, | |
| "logps/chosen": -526.9544067382812, | |
| "logps/rejected": -622.9330444335938, | |
| "loss": 0.1859, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.06152445822954178, | |
| "rewards/margins": 0.049809303134679794, | |
| "rewards/rejected": -0.11133376508951187, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.3014132288708209e-06, | |
| "logits/chosen": -0.04959375411272049, | |
| "logits/rejected": -0.12484890222549438, | |
| "logps/chosen": -564.7056274414062, | |
| "logps/rejected": -604.1170654296875, | |
| "loss": 0.2058, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.05899285152554512, | |
| "rewards/margins": 0.032194919884204865, | |
| "rewards/rejected": -0.09118776023387909, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.2607669951176549e-06, | |
| "logits/chosen": -0.046819113194942474, | |
| "logits/rejected": -0.02196955308318138, | |
| "logps/chosen": -544.0050048828125, | |
| "logps/rejected": -577.3683471679688, | |
| "loss": 0.1872, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.0536922812461853, | |
| "rewards/margins": 0.038987286388874054, | |
| "rewards/rejected": -0.09267956763505936, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.2205505055416891e-06, | |
| "logits/chosen": -0.07024160772562027, | |
| "logits/rejected": -0.004563197493553162, | |
| "logps/chosen": -593.686279296875, | |
| "logps/rejected": -663.3975830078125, | |
| "loss": 0.2018, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.05577061325311661, | |
| "rewards/margins": 0.04607420414686203, | |
| "rewards/rejected": -0.10184482485055923, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.1807777065131002e-06, | |
| "logits/chosen": -0.07550742477178574, | |
| "logits/rejected": -0.007421135902404785, | |
| "logps/chosen": -610.5904541015625, | |
| "logps/rejected": -628.4864501953125, | |
| "loss": 0.2139, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -0.06365049630403519, | |
| "rewards/margins": 0.018766289576888084, | |
| "rewards/rejected": -0.08241678774356842, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.1414623905380012e-06, | |
| "logits/chosen": -0.020526980981230736, | |
| "logits/rejected": -0.038831211626529694, | |
| "logps/chosen": -613.1051635742188, | |
| "logps/rejected": -662.9939575195312, | |
| "loss": 0.1844, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.057425010949373245, | |
| "rewards/margins": 0.03942105546593666, | |
| "rewards/rejected": -0.09684605151414871, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.1026181914754388e-06, | |
| "logits/chosen": -0.05669023469090462, | |
| "logits/rejected": -0.005404030904173851, | |
| "logps/chosen": -522.0977783203125, | |
| "logps/rejected": -629.5926513671875, | |
| "loss": 0.1752, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.0492081418633461, | |
| "rewards/margins": 0.0713469535112381, | |
| "rewards/rejected": -0.1205551028251648, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.0642585798094136e-06, | |
| "logits/chosen": -0.03868585079908371, | |
| "logits/rejected": -0.15266025066375732, | |
| "logps/chosen": -537.9027099609375, | |
| "logps/rejected": -588.400146484375, | |
| "loss": 0.1912, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.04704970866441727, | |
| "rewards/margins": 0.03578130528330803, | |
| "rewards/rejected": -0.0828310176730156, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.0263968579775522e-06, | |
| "logits/chosen": -0.039743565022945404, | |
| "logits/rejected": -0.05752366781234741, | |
| "logps/chosen": -541.8310546875, | |
| "logps/rejected": -601.9227294921875, | |
| "loss": 0.2017, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.04554866999387741, | |
| "rewards/margins": 0.043122462928295135, | |
| "rewards/rejected": -0.08867113292217255, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.89046155758058e-07, | |
| "logits/chosen": -0.01734338328242302, | |
| "logits/rejected": -0.03730212524533272, | |
| "logps/chosen": -536.6981201171875, | |
| "logps/rejected": -599.0525512695312, | |
| "loss": 0.1948, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.05502275377511978, | |
| "rewards/margins": 0.03183166682720184, | |
| "rewards/rejected": -0.08685442060232162, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.52219425716534e-07, | |
| "logits/chosen": -0.0649690106511116, | |
| "logits/rejected": -0.030281294137239456, | |
| "logps/chosen": -510.13037109375, | |
| "logps/rejected": -563.4937133789062, | |
| "loss": 0.1828, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.04460709169507027, | |
| "rewards/margins": 0.03472636267542839, | |
| "rewards/rejected": -0.07933344691991806, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 9.15929438714262e-07, | |
| "logits/chosen": -0.039043981581926346, | |
| "logits/rejected": -0.025227446109056473, | |
| "logps/chosen": -558.3739013671875, | |
| "logps/rejected": -604.859375, | |
| "loss": 0.1936, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.042985010892152786, | |
| "rewards/margins": 0.040365200489759445, | |
| "rewards/rejected": -0.08335021138191223, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.801887794794911e-07, | |
| "logits/chosen": -0.044082947075366974, | |
| "logits/rejected": 0.006080737803131342, | |
| "logps/chosen": -583.2000122070312, | |
| "logps/rejected": -629.2266845703125, | |
| "loss": 0.1878, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.042018450796604156, | |
| "rewards/margins": 0.03190966323018074, | |
| "rewards/rejected": -0.0739281103014946, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.450098422432787e-07, | |
| "logits/chosen": -0.014850256033241749, | |
| "logits/rejected": -0.14079946279525757, | |
| "logps/chosen": -530.8626708984375, | |
| "logps/rejected": -583.5850830078125, | |
| "loss": 0.1933, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.0459921732544899, | |
| "rewards/margins": 0.03213239461183548, | |
| "rewards/rejected": -0.07812456786632538, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.104048264413858e-07, | |
| "logits/chosen": -0.10342959314584732, | |
| "logits/rejected": 0.01154373586177826, | |
| "logps/chosen": -565.1043701171875, | |
| "logps/rejected": -615.1676025390625, | |
| "loss": 0.1802, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.0521150641143322, | |
| "rewards/margins": 0.03459666296839714, | |
| "rewards/rejected": -0.08671173453330994, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.763857324837321e-07, | |
| "logits/chosen": -0.025015175342559814, | |
| "logits/rejected": -0.002726233098655939, | |
| "logps/chosen": -575.7015380859375, | |
| "logps/rejected": -641.2514038085938, | |
| "loss": 0.1856, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.04756324738264084, | |
| "rewards/margins": 0.04533480852842331, | |
| "rewards/rejected": -0.09289805591106415, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.429643575928605e-07, | |
| "logits/chosen": 0.006048756651580334, | |
| "logits/rejected": -0.0436415895819664, | |
| "logps/chosen": -544.4678955078125, | |
| "logps/rejected": -597.9765625, | |
| "loss": 0.1921, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.0511322021484375, | |
| "rewards/margins": 0.02459198795258999, | |
| "rewards/rejected": -0.07572419196367264, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.101522917128709e-07, | |
| "logits/chosen": -0.06234214827418327, | |
| "logits/rejected": 0.00918728206306696, | |
| "logps/chosen": -555.4013061523438, | |
| "logps/rejected": -603.6314697265625, | |
| "loss": 0.195, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.05011584237217903, | |
| "rewards/margins": 0.03172696754336357, | |
| "rewards/rejected": -0.0818428173661232, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.779609134902312e-07, | |
| "logits/chosen": -0.043599605560302734, | |
| "logits/rejected": -0.10574928671121597, | |
| "logps/chosen": -516.0931396484375, | |
| "logps/rejected": -598.8382568359375, | |
| "loss": 0.1858, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.047776833176612854, | |
| "rewards/margins": 0.048279162496328354, | |
| "rewards/rejected": -0.09605599194765091, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.464013863278629e-07, | |
| "logits/chosen": -0.001982520567253232, | |
| "logits/rejected": -0.0532495453953743, | |
| "logps/chosen": -522.6332397460938, | |
| "logps/rejected": -572.3375244140625, | |
| "loss": 0.1889, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.04353585094213486, | |
| "rewards/margins": 0.04176926985383034, | |
| "rewards/rejected": -0.0853051170706749, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.154846545138696e-07, | |
| "logits/chosen": -0.05790393799543381, | |
| "logits/rejected": -0.00911637395620346, | |
| "logps/chosen": -572.7697143554688, | |
| "logps/rejected": -623.4581909179688, | |
| "loss": 0.1777, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.049880485981702805, | |
| "rewards/margins": 0.043653905391693115, | |
| "rewards/rejected": -0.09353439509868622, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.852214394262515e-07, | |
| "logits/chosen": -0.012275640852749348, | |
| "logits/rejected": -0.04724960774183273, | |
| "logps/chosen": -537.2650756835938, | |
| "logps/rejected": -643.6207275390625, | |
| "loss": 0.1696, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.03831138834357262, | |
| "rewards/margins": 0.057165808975696564, | |
| "rewards/rejected": -0.09547720104455948, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.556222358149191e-07, | |
| "logits/chosen": -0.03172417730093002, | |
| "logits/rejected": -0.1095399409532547, | |
| "logps/chosen": -506.0896911621094, | |
| "logps/rejected": -564.1954345703125, | |
| "loss": 0.1888, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.047778304666280746, | |
| "rewards/margins": 0.02731451950967312, | |
| "rewards/rejected": -0.07509282976388931, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.266973081622992e-07, | |
| "logits/chosen": -0.079510398209095, | |
| "logits/rejected": -0.025600697845220566, | |
| "logps/chosen": -502.23529052734375, | |
| "logps/rejected": -601.0836181640625, | |
| "loss": 0.1855, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.05057615786790848, | |
| "rewards/margins": 0.03538522124290466, | |
| "rewards/rejected": -0.08596137166023254, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.984566871237942e-07, | |
| "logits/chosen": -0.05354876443743706, | |
| "logits/rejected": -0.003774791955947876, | |
| "logps/chosen": -595.66357421875, | |
| "logps/rejected": -653.7220458984375, | |
| "loss": 0.1824, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.0476485937833786, | |
| "rewards/margins": 0.04878585413098335, | |
| "rewards/rejected": -0.09643445163965225, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.709101660493251e-07, | |
| "logits/chosen": -0.051042985171079636, | |
| "logits/rejected": -0.04658942669630051, | |
| "logps/chosen": -561.7081298828125, | |
| "logps/rejected": -605.6749877929688, | |
| "loss": 0.1711, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.04670659080147743, | |
| "rewards/margins": 0.05042141675949097, | |
| "rewards/rejected": -0.0971280112862587, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.440672975871743e-07, | |
| "logits/chosen": -0.05640207976102829, | |
| "logits/rejected": -0.034627966582775116, | |
| "logps/chosen": -542.857421875, | |
| "logps/rejected": -607.437744140625, | |
| "loss": 0.1882, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.04426370561122894, | |
| "rewards/margins": 0.04681876674294472, | |
| "rewards/rejected": -0.09108246862888336, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.1793739037129134e-07, | |
| "logits/chosen": -0.02437787875533104, | |
| "logits/rejected": -0.011664156801998615, | |
| "logps/chosen": -626.6923828125, | |
| "logps/rejected": -659.685546875, | |
| "loss": 0.1894, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.052075475454330444, | |
| "rewards/margins": 0.04108622670173645, | |
| "rewards/rejected": -0.0931617021560669, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.9252950579322405e-07, | |
| "logits/chosen": -0.03457965329289436, | |
| "logits/rejected": -0.04087120667099953, | |
| "logps/chosen": -503.4874572753906, | |
| "logps/rejected": -566.9150390625, | |
| "loss": 0.186, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.0481523722410202, | |
| "rewards/margins": 0.04138932749629021, | |
| "rewards/rejected": -0.08954168856143951, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.6785245485978864e-07, | |
| "logits/chosen": -0.12483050674200058, | |
| "logits/rejected": -0.037600237876176834, | |
| "logps/chosen": -563.97314453125, | |
| "logps/rejected": -627.5960693359375, | |
| "loss": 0.2002, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.06187894195318222, | |
| "rewards/margins": 0.03723733127117157, | |
| "rewards/rejected": -0.0991162657737732, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.43914795137566e-07, | |
| "logits/chosen": -0.05473123863339424, | |
| "logits/rejected": -0.07926475256681442, | |
| "logps/chosen": -551.5614624023438, | |
| "logps/rejected": -604.369384765625, | |
| "loss": 0.184, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.046952735632658005, | |
| "rewards/margins": 0.046221982687711716, | |
| "rewards/rejected": -0.09317471832036972, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.207248277852901e-07, | |
| "logits/chosen": -0.1253429353237152, | |
| "logits/rejected": -0.08013667166233063, | |
| "logps/chosen": -565.1073608398438, | |
| "logps/rejected": -605.0519409179688, | |
| "loss": 0.1981, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.05863552540540695, | |
| "rewards/margins": 0.026683971285820007, | |
| "rewards/rejected": -0.08531949669122696, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.9829059467515074e-07, | |
| "logits/chosen": -0.05738731101155281, | |
| "logits/rejected": -0.054315079003572464, | |
| "logps/chosen": -526.6010131835938, | |
| "logps/rejected": -576.3739013671875, | |
| "loss": 0.1928, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.04895392805337906, | |
| "rewards/margins": 0.04191981628537178, | |
| "rewards/rejected": -0.09087374061346054, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.766198756040153e-07, | |
| "logits/chosen": -0.02228875830769539, | |
| "logits/rejected": -0.031645432114601135, | |
| "logps/chosen": -580.3348388671875, | |
| "logps/rejected": -626.4335327148438, | |
| "loss": 0.1859, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.05218944698572159, | |
| "rewards/margins": 0.04899830371141434, | |
| "rewards/rejected": -0.10118775069713593, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.5572018559553155e-07, | |
| "logits/chosen": 0.026733970269560814, | |
| "logits/rejected": -0.07179022580385208, | |
| "logps/chosen": -546.0059814453125, | |
| "logps/rejected": -609.9427490234375, | |
| "loss": 0.1785, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.05384701490402222, | |
| "rewards/margins": 0.04153439775109291, | |
| "rewards/rejected": -0.09538142383098602, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.3559877229404864e-07, | |
| "logits/chosen": -0.10137800127267838, | |
| "logits/rejected": -0.06986267119646072, | |
| "logps/chosen": -515.6650390625, | |
| "logps/rejected": -605.7598876953125, | |
| "loss": 0.182, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.050880689173936844, | |
| "rewards/margins": 0.054004967212677, | |
| "rewards/rejected": -0.10488566011190414, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.1626261345126576e-07, | |
| "logits/chosen": -0.05389503762125969, | |
| "logits/rejected": -0.05351116135716438, | |
| "logps/chosen": -596.9784545898438, | |
| "logps/rejected": -662.634521484375, | |
| "loss": 0.185, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.06371568143367767, | |
| "rewards/margins": 0.05009468272328377, | |
| "rewards/rejected": -0.11381037533283234, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.9771841450646505e-07, | |
| "logits/chosen": -0.12013034522533417, | |
| "logits/rejected": 0.005744548514485359, | |
| "logps/chosen": -562.4629516601562, | |
| "logps/rejected": -641.6666259765625, | |
| "loss": 0.1958, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.047561004757881165, | |
| "rewards/margins": 0.061145197600126266, | |
| "rewards/rejected": -0.10870621353387833, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.7997260626118758e-07, | |
| "logits/chosen": -0.09952473640441895, | |
| "logits/rejected": -0.06873885542154312, | |
| "logps/chosen": -495.61199951171875, | |
| "logps/rejected": -587.5067138671875, | |
| "loss": 0.1836, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.05300045758485794, | |
| "rewards/margins": 0.050713229924440384, | |
| "rewards/rejected": -0.10371369123458862, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.6303134264914365e-07, | |
| "logits/chosen": -0.025821620598435402, | |
| "logits/rejected": -0.15098163485527039, | |
| "logps/chosen": -559.591064453125, | |
| "logps/rejected": -673.056640625, | |
| "loss": 0.1711, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.06007053703069687, | |
| "rewards/margins": 0.05461307615041733, | |
| "rewards/rejected": -0.1146836131811142, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.469004986021355e-07, | |
| "logits/chosen": 0.037708066403865814, | |
| "logits/rejected": -0.11987550556659698, | |
| "logps/chosen": -585.4603881835938, | |
| "logps/rejected": -654.9968872070312, | |
| "loss": 0.1686, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.05478737875819206, | |
| "rewards/margins": 0.05192810297012329, | |
| "rewards/rejected": -0.10671547800302505, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.315856680127367e-07, | |
| "logits/chosen": -0.023753302171826363, | |
| "logits/rejected": -0.022690903395414352, | |
| "logps/chosen": -563.29345703125, | |
| "logps/rejected": -643.427490234375, | |
| "loss": 0.1791, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.05222300440073013, | |
| "rewards/margins": 0.05298025533556938, | |
| "rewards/rejected": -0.10520325601100922, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.1709216179442817e-07, | |
| "logits/chosen": -0.07842310518026352, | |
| "logits/rejected": -0.06540755927562714, | |
| "logps/chosen": -601.8659057617188, | |
| "logps/rejected": -662.879150390625, | |
| "loss": 0.1923, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.06227841228246689, | |
| "rewards/margins": 0.040394991636276245, | |
| "rewards/rejected": -0.10267341136932373, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.0342500603986421e-07, | |
| "logits/chosen": -0.05102648586034775, | |
| "logits/rejected": -0.06519290059804916, | |
| "logps/chosen": -558.7926025390625, | |
| "logps/rejected": -606.8863525390625, | |
| "loss": 0.1921, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.0554363913834095, | |
| "rewards/margins": 0.03473091870546341, | |
| "rewards/rejected": -0.09016730636358261, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 9.058894027791643e-08, | |
| "logits/chosen": -0.005893229506909847, | |
| "logits/rejected": -0.06061660498380661, | |
| "logps/chosen": -520.56201171875, | |
| "logps/rejected": -629.1943969726562, | |
| "loss": 0.1805, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.0561298243701458, | |
| "rewards/margins": 0.04909069091081619, | |
| "rewards/rejected": -0.10522051900625229, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.858841583008592e-08, | |
| "logits/chosen": -0.02195947803556919, | |
| "logits/rejected": -0.0024458705447614193, | |
| "logps/chosen": -613.8434448242188, | |
| "logps/rejected": -640.0768432617188, | |
| "loss": 0.2113, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.06373412162065506, | |
| "rewards/margins": 0.03569970279932022, | |
| "rewards/rejected": -0.09943382441997528, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.742759426686313e-08, | |
| "logits/chosen": -0.05320558696985245, | |
| "logits/rejected": -0.048701487481594086, | |
| "logps/chosen": -581.4453125, | |
| "logps/rejected": -626.544677734375, | |
| "loss": 0.1786, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.058798860758543015, | |
| "rewards/margins": 0.04494379833340645, | |
| "rewards/rejected": -0.10374267399311066, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 5.7110345964571104e-08, | |
| "logits/chosen": -0.06332091242074966, | |
| "logits/rejected": -0.0014598838752135634, | |
| "logps/chosen": -652.0838623046875, | |
| "logps/rejected": -733.8502807617188, | |
| "loss": 0.1778, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.057877421379089355, | |
| "rewards/margins": 0.06819779425859451, | |
| "rewards/rejected": -0.12607520818710327, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.764024876318357e-08, | |
| "logits/chosen": -0.10024017095565796, | |
| "logits/rejected": 0.01046714186668396, | |
| "logps/chosen": -522.90185546875, | |
| "logps/rejected": -571.7759399414062, | |
| "loss": 0.2054, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.06100086122751236, | |
| "rewards/margins": 0.03126870468258858, | |
| "rewards/rejected": -0.09226956218481064, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.902058672559633e-08, | |
| "logits/chosen": 0.03307167440652847, | |
| "logits/rejected": -0.036007530987262726, | |
| "logps/chosen": -591.3616943359375, | |
| "logps/rejected": -665.3886108398438, | |
| "loss": 0.1901, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.05448399856686592, | |
| "rewards/margins": 0.04371560364961624, | |
| "rewards/rejected": -0.09819960594177246, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.125434899876933e-08, | |
| "logits/chosen": 0.0261197779327631, | |
| "logits/rejected": -0.10165154933929443, | |
| "logps/chosen": -611.7322387695312, | |
| "logps/rejected": -695.073974609375, | |
| "loss": 0.197, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.060929931700229645, | |
| "rewards/margins": 0.049871496856212616, | |
| "rewards/rejected": -0.11080142110586166, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.4344228777145873e-08, | |
| "logits/chosen": -0.10922287404537201, | |
| "logits/rejected": -0.036579955369234085, | |
| "logps/chosen": -538.944091796875, | |
| "logps/rejected": -600.6095581054688, | |
| "loss": 0.1917, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.05488158017396927, | |
| "rewards/margins": 0.04075679928064346, | |
| "rewards/rejected": -0.09563836455345154, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.829262236869772e-08, | |
| "logits/chosen": 0.006532920990139246, | |
| "logits/rejected": -0.0661952868103981, | |
| "logps/chosen": -537.886962890625, | |
| "logps/rejected": -634.1477661132812, | |
| "loss": 0.2016, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.05655653029680252, | |
| "rewards/margins": 0.06689898669719696, | |
| "rewards/rejected": -0.12345552444458008, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3101628363929586e-08, | |
| "logits/chosen": -0.0753311961889267, | |
| "logits/rejected": -0.013547460548579693, | |
| "logps/chosen": -561.2381591796875, | |
| "logps/rejected": -576.2022094726562, | |
| "loss": 0.1924, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.058451224118471146, | |
| "rewards/margins": 0.025429999455809593, | |
| "rewards/rejected": -0.08388122171163559, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.773046908123195e-09, | |
| "logits/chosen": -0.022709984332323074, | |
| "logits/rejected": -0.028875475749373436, | |
| "logps/chosen": -606.6187744140625, | |
| "logps/rejected": -637.9049072265625, | |
| "loss": 0.198, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.072281613945961, | |
| "rewards/margins": 0.030539941042661667, | |
| "rewards/rejected": -0.10282155126333237, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 5.308379077080817e-09, | |
| "logits/chosen": 0.0019970417488366365, | |
| "logits/rejected": -0.02689496800303459, | |
| "logps/chosen": -602.9913940429688, | |
| "logps/rejected": -666.972412109375, | |
| "loss": 0.196, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.057968251407146454, | |
| "rewards/margins": 0.04878325015306473, | |
| "rewards/rejected": -0.10675151646137238, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.7088263565760996e-09, | |
| "logits/chosen": -0.10940905660390854, | |
| "logits/rejected": -0.03552763909101486, | |
| "logps/chosen": -565.7769775390625, | |
| "logps/rejected": -600.9927368164062, | |
| "loss": 0.1964, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.04456068575382233, | |
| "rewards/margins": 0.03636551275849342, | |
| "rewards/rejected": -0.08092619478702545, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.752902257023633e-10, | |
| "logits/chosen": -0.02857157588005066, | |
| "logits/rejected": 0.05758611112833023, | |
| "logps/chosen": -594.5465087890625, | |
| "logps/rejected": -621.6343994140625, | |
| "loss": 0.1832, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.05167108029127121, | |
| "rewards/margins": 0.037888895720243454, | |
| "rewards/rejected": -0.08955998718738556, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.083718442532189e-10, | |
| "logits/chosen": -0.07677461951971054, | |
| "logits/rejected": 0.008063477464020252, | |
| "logps/chosen": -562.1439208984375, | |
| "logps/rejected": -642.2802734375, | |
| "loss": 0.1852, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.0575174018740654, | |
| "rewards/margins": 0.049335263669490814, | |
| "rewards/rejected": -0.10685265064239502, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 1875, | |
| "total_flos": 0.0, | |
| "train_loss": 0.19300631308555602, | |
| "train_runtime": 11612.5974, | |
| "train_samples_per_second": 2.583, | |
| "train_steps_per_second": 0.161 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1875, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |