| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 100, | |
| "global_step": 4164, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007204610951008645, | |
| "grad_norm": 14.58157977905889, | |
| "learning_rate": 1.199040767386091e-10, | |
| "logits/chosen": -1.901450514793396, | |
| "logits/rejected": -1.9076323509216309, | |
| "logps/chosen": -0.8524526953697205, | |
| "logps/rejected": -0.9626365900039673, | |
| "loss": 1.1927, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -1.704905390739441, | |
| "rewards/margins": 0.22036786377429962, | |
| "rewards/rejected": -1.9252731800079346, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007204610951008645, | |
| "grad_norm": 17.709463159121455, | |
| "learning_rate": 1.199040767386091e-09, | |
| "logits/chosen": -2.020684242248535, | |
| "logits/rejected": -2.0064282417297363, | |
| "logps/chosen": -1.0048482418060303, | |
| "logps/rejected": -1.1098697185516357, | |
| "loss": 1.216, | |
| "rewards/accuracies": 0.5208333134651184, | |
| "rewards/chosen": -2.0096964836120605, | |
| "rewards/margins": 0.21004274487495422, | |
| "rewards/rejected": -2.2197394371032715, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01440922190201729, | |
| "grad_norm": 22.640302051500377, | |
| "learning_rate": 2.398081534772182e-09, | |
| "logits/chosen": -2.021089792251587, | |
| "logits/rejected": -2.0176689624786377, | |
| "logps/chosen": -1.0516496896743774, | |
| "logps/rejected": -1.1834802627563477, | |
| "loss": 1.1858, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.103299379348755, | |
| "rewards/margins": 0.26366108655929565, | |
| "rewards/rejected": -2.3669605255126953, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021613832853025938, | |
| "grad_norm": 17.8606028438409, | |
| "learning_rate": 3.597122302158273e-09, | |
| "logits/chosen": -1.9866092205047607, | |
| "logits/rejected": -1.9793494939804077, | |
| "logps/chosen": -1.0540482997894287, | |
| "logps/rejected": -1.1519711017608643, | |
| "loss": 1.2346, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1080965995788574, | |
| "rewards/margins": 0.19584545493125916, | |
| "rewards/rejected": -2.3039422035217285, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02881844380403458, | |
| "grad_norm": 19.245572130250604, | |
| "learning_rate": 4.796163069544364e-09, | |
| "logits/chosen": -2.0317888259887695, | |
| "logits/rejected": -2.031811475753784, | |
| "logps/chosen": -1.0351777076721191, | |
| "logps/rejected": -1.136722207069397, | |
| "loss": 1.2355, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.0703554153442383, | |
| "rewards/margins": 0.20308911800384521, | |
| "rewards/rejected": -2.273444414138794, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03602305475504323, | |
| "grad_norm": 14.943806509066846, | |
| "learning_rate": 5.995203836930456e-09, | |
| "logits/chosen": -1.9625627994537354, | |
| "logits/rejected": -1.9631847143173218, | |
| "logps/chosen": -0.9414892196655273, | |
| "logps/rejected": -1.007533311843872, | |
| "loss": 1.2547, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.8829784393310547, | |
| "rewards/margins": 0.13208839297294617, | |
| "rewards/rejected": -2.015066623687744, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.043227665706051875, | |
| "grad_norm": 21.528231741291215, | |
| "learning_rate": 7.194244604316546e-09, | |
| "logits/chosen": -2.033930778503418, | |
| "logits/rejected": -2.0294690132141113, | |
| "logps/chosen": -1.0896106958389282, | |
| "logps/rejected": -1.1459602117538452, | |
| "loss": 1.2679, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -2.1792213916778564, | |
| "rewards/margins": 0.1126992255449295, | |
| "rewards/rejected": -2.2919204235076904, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05043227665706052, | |
| "grad_norm": 20.70296936549822, | |
| "learning_rate": 8.393285371702639e-09, | |
| "logits/chosen": -2.0241129398345947, | |
| "logits/rejected": -2.0117270946502686, | |
| "logps/chosen": -1.1098978519439697, | |
| "logps/rejected": -1.204820156097412, | |
| "loss": 1.2271, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -2.2197957038879395, | |
| "rewards/margins": 0.1898445188999176, | |
| "rewards/rejected": -2.409640312194824, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.05763688760806916, | |
| "grad_norm": 24.40623296093575, | |
| "learning_rate": 9.592326139088728e-09, | |
| "logits/chosen": -2.0398144721984863, | |
| "logits/rejected": -2.036891222000122, | |
| "logps/chosen": -1.1656566858291626, | |
| "logps/rejected": -1.237831473350525, | |
| "loss": 1.2527, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.331313371658325, | |
| "rewards/margins": 0.14434944093227386, | |
| "rewards/rejected": -2.47566294670105, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06484149855907781, | |
| "grad_norm": 15.525751311455734, | |
| "learning_rate": 1.0791366906474819e-08, | |
| "logits/chosen": -2.0057613849639893, | |
| "logits/rejected": -2.0072615146636963, | |
| "logps/chosen": -1.0418776273727417, | |
| "logps/rejected": -1.1488852500915527, | |
| "loss": 1.215, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.0837552547454834, | |
| "rewards/margins": 0.21401505172252655, | |
| "rewards/rejected": -2.2977705001831055, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07204610951008646, | |
| "grad_norm": 19.01739570575657, | |
| "learning_rate": 1.1990407673860912e-08, | |
| "logits/chosen": -2.0440549850463867, | |
| "logits/rejected": -2.038007974624634, | |
| "logps/chosen": -1.0073726177215576, | |
| "logps/rejected": -1.114424467086792, | |
| "loss": 1.2172, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0147452354431152, | |
| "rewards/margins": 0.21410349011421204, | |
| "rewards/rejected": -2.228848934173584, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0792507204610951, | |
| "grad_norm": 16.468864603689383, | |
| "learning_rate": 1.3189448441247003e-08, | |
| "logits/chosen": -1.986783742904663, | |
| "logits/rejected": -1.975547194480896, | |
| "logps/chosen": -1.0294089317321777, | |
| "logps/rejected": -1.1291263103485107, | |
| "loss": 1.2279, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0588178634643555, | |
| "rewards/margins": 0.19943459331989288, | |
| "rewards/rejected": -2.2582526206970215, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08645533141210375, | |
| "grad_norm": 18.27069220463476, | |
| "learning_rate": 1.4388489208633092e-08, | |
| "logits/chosen": -1.9731948375701904, | |
| "logits/rejected": -1.9713401794433594, | |
| "logps/chosen": -0.9640307426452637, | |
| "logps/rejected": -1.0653537511825562, | |
| "loss": 1.2087, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.9280614852905273, | |
| "rewards/margins": 0.2026461362838745, | |
| "rewards/rejected": -2.1307075023651123, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0936599423631124, | |
| "grad_norm": 17.232187953156046, | |
| "learning_rate": 1.5587529976019183e-08, | |
| "logits/chosen": -2.066575527191162, | |
| "logits/rejected": -2.065995931625366, | |
| "logps/chosen": -1.0801920890808105, | |
| "logps/rejected": -1.1521753072738647, | |
| "loss": 1.2549, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.160384178161621, | |
| "rewards/margins": 0.14396657049655914, | |
| "rewards/rejected": -2.3043506145477295, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.10086455331412104, | |
| "grad_norm": 20.847348575081657, | |
| "learning_rate": 1.6786570743405277e-08, | |
| "logits/chosen": -1.9832985401153564, | |
| "logits/rejected": -1.9769630432128906, | |
| "logps/chosen": -0.9781940579414368, | |
| "logps/rejected": -1.122657060623169, | |
| "loss": 1.1694, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9563881158828735, | |
| "rewards/margins": 0.2889261543750763, | |
| "rewards/rejected": -2.245314121246338, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.10806916426512968, | |
| "grad_norm": 19.95238793204191, | |
| "learning_rate": 1.7985611510791365e-08, | |
| "logits/chosen": -1.9963840246200562, | |
| "logits/rejected": -1.9920928478240967, | |
| "logps/chosen": -1.0187867879867554, | |
| "logps/rejected": -1.136918306350708, | |
| "loss": 1.2067, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.0375735759735107, | |
| "rewards/margins": 0.23626303672790527, | |
| "rewards/rejected": -2.273836612701416, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.11527377521613832, | |
| "grad_norm": 17.4507491502089, | |
| "learning_rate": 1.9184652278177456e-08, | |
| "logits/chosen": -2.00455904006958, | |
| "logits/rejected": -1.9985454082489014, | |
| "logps/chosen": -0.9479260444641113, | |
| "logps/rejected": -1.0970423221588135, | |
| "loss": 1.1509, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -1.8958520889282227, | |
| "rewards/margins": 0.2982328534126282, | |
| "rewards/rejected": -2.194084644317627, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.12247838616714697, | |
| "grad_norm": 22.64495005377011, | |
| "learning_rate": 2.038369304556355e-08, | |
| "logits/chosen": -2.0030527114868164, | |
| "logits/rejected": -1.995448350906372, | |
| "logps/chosen": -1.0368740558624268, | |
| "logps/rejected": -1.1604634523391724, | |
| "loss": 1.2057, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.0737481117248535, | |
| "rewards/margins": 0.24717874825000763, | |
| "rewards/rejected": -2.3209269046783447, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.12968299711815562, | |
| "grad_norm": 23.590437364971006, | |
| "learning_rate": 2.1582733812949638e-08, | |
| "logits/chosen": -2.0346579551696777, | |
| "logits/rejected": -2.027749538421631, | |
| "logps/chosen": -1.020750641822815, | |
| "logps/rejected": -1.1084620952606201, | |
| "loss": 1.2476, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.04150128364563, | |
| "rewards/margins": 0.17542308568954468, | |
| "rewards/rejected": -2.2169241905212402, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.13688760806916425, | |
| "grad_norm": 22.966874261128403, | |
| "learning_rate": 2.278177458033573e-08, | |
| "logits/chosen": -2.073704719543457, | |
| "logits/rejected": -2.0714824199676514, | |
| "logps/chosen": -0.9697921872138977, | |
| "logps/rejected": -1.065453290939331, | |
| "loss": 1.212, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.9395843744277954, | |
| "rewards/margins": 0.19132229685783386, | |
| "rewards/rejected": -2.130906581878662, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.1440922190201729, | |
| "grad_norm": 22.638490791764895, | |
| "learning_rate": 2.3980815347721823e-08, | |
| "logits/chosen": -2.0427424907684326, | |
| "logits/rejected": -2.0397419929504395, | |
| "logps/chosen": -1.0259110927581787, | |
| "logps/rejected": -1.1529022455215454, | |
| "loss": 1.1871, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0518221855163574, | |
| "rewards/margins": 0.2539823353290558, | |
| "rewards/rejected": -2.305804491043091, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15129682997118155, | |
| "grad_norm": 21.113736148839788, | |
| "learning_rate": 2.517985611510791e-08, | |
| "logits/chosen": -2.0403473377227783, | |
| "logits/rejected": -2.037600040435791, | |
| "logps/chosen": -1.0739350318908691, | |
| "logps/rejected": -1.150781273841858, | |
| "loss": 1.2504, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.1478700637817383, | |
| "rewards/margins": 0.15369237959384918, | |
| "rewards/rejected": -2.301562547683716, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.1585014409221902, | |
| "grad_norm": 15.482070000655302, | |
| "learning_rate": 2.6378896882494006e-08, | |
| "logits/chosen": -1.9863160848617554, | |
| "logits/rejected": -1.982267141342163, | |
| "logps/chosen": -1.0080206394195557, | |
| "logps/rejected": -1.176837682723999, | |
| "loss": 1.1505, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.0160412788391113, | |
| "rewards/margins": 0.3376340866088867, | |
| "rewards/rejected": -2.353675365447998, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.16570605187319884, | |
| "grad_norm": 17.014637756082593, | |
| "learning_rate": 2.7577937649880097e-08, | |
| "logits/chosen": -2.021378993988037, | |
| "logits/rejected": -2.021695613861084, | |
| "logps/chosen": -1.0124410390853882, | |
| "logps/rejected": -1.12635338306427, | |
| "loss": 1.2019, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.0248820781707764, | |
| "rewards/margins": 0.22782447934150696, | |
| "rewards/rejected": -2.25270676612854, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.1729106628242075, | |
| "grad_norm": 22.32772580016105, | |
| "learning_rate": 2.8776978417266184e-08, | |
| "logits/chosen": -2.0529181957244873, | |
| "logits/rejected": -2.0477967262268066, | |
| "logps/chosen": -1.0616161823272705, | |
| "logps/rejected": -1.1394503116607666, | |
| "loss": 1.2614, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.123232364654541, | |
| "rewards/margins": 0.15566802024841309, | |
| "rewards/rejected": -2.278900623321533, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.18011527377521613, | |
| "grad_norm": 19.079728631088813, | |
| "learning_rate": 2.997601918465228e-08, | |
| "logits/chosen": -1.9696033000946045, | |
| "logits/rejected": -1.9657630920410156, | |
| "logps/chosen": -1.0835182666778564, | |
| "logps/rejected": -1.1734166145324707, | |
| "loss": 1.2393, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.167036533355713, | |
| "rewards/margins": 0.17979690432548523, | |
| "rewards/rejected": -2.3468332290649414, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1873198847262248, | |
| "grad_norm": 21.30398020890557, | |
| "learning_rate": 3.1175059952038366e-08, | |
| "logits/chosen": -1.9843509197235107, | |
| "logits/rejected": -1.9924278259277344, | |
| "logps/chosen": -1.1062877178192139, | |
| "logps/rejected": -1.2165796756744385, | |
| "loss": 1.2142, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.2125754356384277, | |
| "rewards/margins": 0.22058391571044922, | |
| "rewards/rejected": -2.433159351348877, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.19452449567723343, | |
| "grad_norm": 20.993622960377618, | |
| "learning_rate": 3.237410071942446e-08, | |
| "logits/chosen": -2.0651376247406006, | |
| "logits/rejected": -2.0571722984313965, | |
| "logps/chosen": -1.0719540119171143, | |
| "logps/rejected": -1.2004284858703613, | |
| "loss": 1.181, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.1439080238342285, | |
| "rewards/margins": 0.2569490075111389, | |
| "rewards/rejected": -2.4008569717407227, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.2017291066282421, | |
| "grad_norm": 25.067055659781758, | |
| "learning_rate": 3.3573141486810555e-08, | |
| "logits/chosen": -2.014195680618286, | |
| "logits/rejected": -2.012540102005005, | |
| "logps/chosen": -0.935396671295166, | |
| "logps/rejected": -1.049852967262268, | |
| "loss": 1.1977, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -1.870793342590332, | |
| "rewards/margins": 0.2289123237133026, | |
| "rewards/rejected": -2.099705934524536, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.20893371757925072, | |
| "grad_norm": 21.777264205916122, | |
| "learning_rate": 3.477218225419664e-08, | |
| "logits/chosen": -2.044172763824463, | |
| "logits/rejected": -2.0461270809173584, | |
| "logps/chosen": -1.0135643482208252, | |
| "logps/rejected": -1.1082309484481812, | |
| "loss": 1.2343, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0271286964416504, | |
| "rewards/margins": 0.18933361768722534, | |
| "rewards/rejected": -2.2164618968963623, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.21613832853025935, | |
| "grad_norm": 20.318543545834533, | |
| "learning_rate": 3.597122302158273e-08, | |
| "logits/chosen": -2.0240025520324707, | |
| "logits/rejected": -2.0156774520874023, | |
| "logps/chosen": -1.0902036428451538, | |
| "logps/rejected": -1.1914021968841553, | |
| "loss": 1.2135, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1804072856903076, | |
| "rewards/margins": 0.20239713788032532, | |
| "rewards/rejected": -2.3828043937683105, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22334293948126802, | |
| "grad_norm": 18.50470861360763, | |
| "learning_rate": 3.717026378896883e-08, | |
| "logits/chosen": -1.9557920694351196, | |
| "logits/rejected": -1.955775260925293, | |
| "logps/chosen": -1.0874634981155396, | |
| "logps/rejected": -1.1727240085601807, | |
| "loss": 1.2381, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.174926996231079, | |
| "rewards/margins": 0.17052076756954193, | |
| "rewards/rejected": -2.3454480171203613, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.23054755043227665, | |
| "grad_norm": 15.935054480540096, | |
| "learning_rate": 3.836930455635491e-08, | |
| "logits/chosen": -2.031646966934204, | |
| "logits/rejected": -2.0232386589050293, | |
| "logps/chosen": -1.0084177255630493, | |
| "logps/rejected": -1.1408658027648926, | |
| "loss": 1.1926, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.0168354511260986, | |
| "rewards/margins": 0.2648962140083313, | |
| "rewards/rejected": -2.281731605529785, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.2377521613832853, | |
| "grad_norm": 15.808626134367197, | |
| "learning_rate": 3.9568345323741003e-08, | |
| "logits/chosen": -2.019885778427124, | |
| "logits/rejected": -2.022150754928589, | |
| "logps/chosen": -1.0463831424713135, | |
| "logps/rejected": -1.069990873336792, | |
| "loss": 1.3364, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -2.092766284942627, | |
| "rewards/margins": 0.04721563309431076, | |
| "rewards/rejected": -2.139981746673584, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.24495677233429394, | |
| "grad_norm": 18.32115617252851, | |
| "learning_rate": 4.07673860911271e-08, | |
| "logits/chosen": -2.0614123344421387, | |
| "logits/rejected": -2.055767297744751, | |
| "logps/chosen": -1.0877503156661987, | |
| "logps/rejected": -1.16796875, | |
| "loss": 1.2366, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.1755006313323975, | |
| "rewards/margins": 0.16043710708618164, | |
| "rewards/rejected": -2.3359375, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.2521613832853026, | |
| "grad_norm": 19.369790564686102, | |
| "learning_rate": 4.1966426858513185e-08, | |
| "logits/chosen": -1.9940099716186523, | |
| "logits/rejected": -1.9883639812469482, | |
| "logps/chosen": -0.9887149930000305, | |
| "logps/rejected": -1.1154861450195312, | |
| "loss": 1.1858, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -1.977429986000061, | |
| "rewards/margins": 0.25354230403900146, | |
| "rewards/rejected": -2.2309722900390625, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.25936599423631124, | |
| "grad_norm": 21.686526135721945, | |
| "learning_rate": 4.3165467625899276e-08, | |
| "logits/chosen": -1.9959064722061157, | |
| "logits/rejected": -1.9917312860488892, | |
| "logps/chosen": -1.0866310596466064, | |
| "logps/rejected": -1.2025970220565796, | |
| "loss": 1.1977, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.173262119293213, | |
| "rewards/margins": 0.23193176090717316, | |
| "rewards/rejected": -2.405194044113159, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.2665706051873199, | |
| "grad_norm": 18.21919999535183, | |
| "learning_rate": 4.4364508393285374e-08, | |
| "logits/chosen": -2.0002856254577637, | |
| "logits/rejected": -2.000253200531006, | |
| "logps/chosen": -1.0520254373550415, | |
| "logps/rejected": -1.180267572402954, | |
| "loss": 1.1778, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.104050874710083, | |
| "rewards/margins": 0.2564844489097595, | |
| "rewards/rejected": -2.360535144805908, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.2737752161383285, | |
| "grad_norm": 16.536106044001812, | |
| "learning_rate": 4.556354916067146e-08, | |
| "logits/chosen": -2.028313398361206, | |
| "logits/rejected": -2.032285213470459, | |
| "logps/chosen": -1.0125257968902588, | |
| "logps/rejected": -1.0858430862426758, | |
| "loss": 1.2682, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.0250515937805176, | |
| "rewards/margins": 0.14663462340831757, | |
| "rewards/rejected": -2.1716861724853516, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.28097982708933716, | |
| "grad_norm": 15.31773608533987, | |
| "learning_rate": 4.676258992805755e-08, | |
| "logits/chosen": -2.0320096015930176, | |
| "logits/rejected": -2.0257675647735596, | |
| "logps/chosen": -1.0224783420562744, | |
| "logps/rejected": -1.1486625671386719, | |
| "loss": 1.1819, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.044956684112549, | |
| "rewards/margins": 0.25236865878105164, | |
| "rewards/rejected": -2.2973251342773438, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.2881844380403458, | |
| "grad_norm": 18.995537958721503, | |
| "learning_rate": 4.796163069544365e-08, | |
| "logits/chosen": -2.034123420715332, | |
| "logits/rejected": -2.034450054168701, | |
| "logps/chosen": -0.9964189529418945, | |
| "logps/rejected": -1.0486726760864258, | |
| "loss": 1.2726, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.992837905883789, | |
| "rewards/margins": 0.10450725257396698, | |
| "rewards/rejected": -2.0973453521728516, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2953890489913545, | |
| "grad_norm": 18.624392586338367, | |
| "learning_rate": 4.916067146282973e-08, | |
| "logits/chosen": -2.0277891159057617, | |
| "logits/rejected": -2.0259571075439453, | |
| "logps/chosen": -1.0748345851898193, | |
| "logps/rejected": -1.1457411050796509, | |
| "loss": 1.262, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -2.1496691703796387, | |
| "rewards/margins": 0.14181289076805115, | |
| "rewards/rejected": -2.2914822101593018, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.3025936599423631, | |
| "grad_norm": 16.76581954495512, | |
| "learning_rate": 4.999992091672379e-08, | |
| "logits/chosen": -2.011078119277954, | |
| "logits/rejected": -2.0153493881225586, | |
| "logps/chosen": -1.0450259447097778, | |
| "logps/rejected": -1.1236448287963867, | |
| "loss": 1.2425, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -2.0900518894195557, | |
| "rewards/margins": 0.15723773837089539, | |
| "rewards/rejected": -2.2472896575927734, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.30979827089337175, | |
| "grad_norm": 17.72733209255425, | |
| "learning_rate": 4.999851500573209e-08, | |
| "logits/chosen": -1.9903459548950195, | |
| "logits/rejected": -1.991233229637146, | |
| "logps/chosen": -1.0592777729034424, | |
| "logps/rejected": -1.0997775793075562, | |
| "loss": 1.3022, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -2.1185555458068848, | |
| "rewards/margins": 0.08099973201751709, | |
| "rewards/rejected": -2.1995551586151123, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.3170028818443804, | |
| "grad_norm": 15.96665018689344, | |
| "learning_rate": 4.999535180235972e-08, | |
| "logits/chosen": -1.990563988685608, | |
| "logits/rejected": -1.9907207489013672, | |
| "logps/chosen": -1.0212013721466064, | |
| "logps/rejected": -1.1435030698776245, | |
| "loss": 1.1959, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.042402744293213, | |
| "rewards/margins": 0.2446034699678421, | |
| "rewards/rejected": -2.287006139755249, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.3242074927953891, | |
| "grad_norm": 17.84897470512453, | |
| "learning_rate": 4.9990431528966836e-08, | |
| "logits/chosen": -2.010443925857544, | |
| "logits/rejected": -2.006673574447632, | |
| "logps/chosen": -1.1450097560882568, | |
| "logps/rejected": -1.1849489212036133, | |
| "loss": 1.3018, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -2.2900195121765137, | |
| "rewards/margins": 0.07987822592258453, | |
| "rewards/rejected": -2.3698978424072266, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.3314121037463977, | |
| "grad_norm": 24.49190807066052, | |
| "learning_rate": 4.9983754531428326e-08, | |
| "logits/chosen": -2.006472110748291, | |
| "logits/rejected": -2.00079083442688, | |
| "logps/chosen": -1.1708580255508423, | |
| "logps/rejected": -1.2872368097305298, | |
| "loss": 1.2012, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.3417160511016846, | |
| "rewards/margins": 0.23275737464427948, | |
| "rewards/rejected": -2.5744736194610596, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.33861671469740634, | |
| "grad_norm": 23.024434569130843, | |
| "learning_rate": 4.997532127910954e-08, | |
| "logits/chosen": -2.0429301261901855, | |
| "logits/rejected": -2.0308475494384766, | |
| "logps/chosen": -1.100434422492981, | |
| "logps/rejected": -1.2019624710083008, | |
| "loss": 1.2198, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.200868844985962, | |
| "rewards/margins": 0.20305626094341278, | |
| "rewards/rejected": -2.4039249420166016, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.345821325648415, | |
| "grad_norm": 21.129827787614413, | |
| "learning_rate": 4.996513236483331e-08, | |
| "logits/chosen": -2.101729154586792, | |
| "logits/rejected": -2.091571569442749, | |
| "logps/chosen": -0.9851749539375305, | |
| "logps/rejected": -1.106676459312439, | |
| "loss": 1.185, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.970349907875061, | |
| "rewards/margins": 0.2430029660463333, | |
| "rewards/rejected": -2.213352918624878, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.3530259365994236, | |
| "grad_norm": 18.94655048736081, | |
| "learning_rate": 4.9953188504838225e-08, | |
| "logits/chosen": -2.0206782817840576, | |
| "logits/rejected": -2.0197720527648926, | |
| "logps/chosen": -0.9880903959274292, | |
| "logps/rejected": -1.1017425060272217, | |
| "loss": 1.1937, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.9761807918548584, | |
| "rewards/margins": 0.2273043841123581, | |
| "rewards/rejected": -2.2034850120544434, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.36023054755043227, | |
| "grad_norm": 18.60846892662722, | |
| "learning_rate": 4.993949053872834e-08, | |
| "logits/chosen": -2.019057035446167, | |
| "logits/rejected": -2.0055313110351562, | |
| "logps/chosen": -1.0131161212921143, | |
| "logps/rejected": -1.139453649520874, | |
| "loss": 1.1821, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0262322425842285, | |
| "rewards/margins": 0.2526749074459076, | |
| "rewards/rejected": -2.278907299041748, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36743515850144093, | |
| "grad_norm": 19.18531858517567, | |
| "learning_rate": 4.9924039429414086e-08, | |
| "logits/chosen": -2.0883572101593018, | |
| "logits/rejected": -2.0818283557891846, | |
| "logps/chosen": -1.0440417528152466, | |
| "logps/rejected": -1.1581791639328003, | |
| "loss": 1.2079, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -2.088083505630493, | |
| "rewards/margins": 0.22827525436878204, | |
| "rewards/rejected": -2.3163583278656006, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.3746397694524496, | |
| "grad_norm": 16.068632795684866, | |
| "learning_rate": 4.990683626304467e-08, | |
| "logits/chosen": -2.010894775390625, | |
| "logits/rejected": -2.0092484951019287, | |
| "logps/chosen": -1.1070988178253174, | |
| "logps/rejected": -1.2031704187393188, | |
| "loss": 1.2198, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.2141976356506348, | |
| "rewards/margins": 0.19214320182800293, | |
| "rewards/rejected": -2.4063408374786377, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.3818443804034582, | |
| "grad_norm": 17.727178124609676, | |
| "learning_rate": 4.9887882248931646e-08, | |
| "logits/chosen": -1.97884202003479, | |
| "logits/rejected": -1.968973159790039, | |
| "logps/chosen": -0.9846093058586121, | |
| "logps/rejected": -1.0614283084869385, | |
| "loss": 1.2503, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.9692186117172241, | |
| "rewards/margins": 0.15363821387290955, | |
| "rewards/rejected": -2.122856616973877, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.38904899135446686, | |
| "grad_norm": 22.67722196494781, | |
| "learning_rate": 4.986717871946393e-08, | |
| "logits/chosen": -2.004068374633789, | |
| "logits/rejected": -1.99717116355896, | |
| "logps/chosen": -1.0308892726898193, | |
| "logps/rejected": -1.1323744058609009, | |
| "loss": 1.2209, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.0617785453796387, | |
| "rewards/margins": 0.20297034084796906, | |
| "rewards/rejected": -2.2647488117218018, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.3962536023054755, | |
| "grad_norm": 17.281352424891857, | |
| "learning_rate": 4.984472713001416e-08, | |
| "logits/chosen": -1.9620494842529297, | |
| "logits/rejected": -1.962517499923706, | |
| "logps/chosen": -1.0005210638046265, | |
| "logps/rejected": -1.0776532888412476, | |
| "loss": 1.2683, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -2.001042127609253, | |
| "rewards/margins": 0.1542646884918213, | |
| "rewards/rejected": -2.155306577682495, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.4034582132564842, | |
| "grad_norm": 17.117159642375974, | |
| "learning_rate": 4.982052905883637e-08, | |
| "logits/chosen": -2.031991481781006, | |
| "logits/rejected": -2.0326719284057617, | |
| "logps/chosen": -1.080214262008667, | |
| "logps/rejected": -1.181120753288269, | |
| "loss": 1.224, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.160428524017334, | |
| "rewards/margins": 0.2018129527568817, | |
| "rewards/rejected": -2.362241506576538, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.4106628242074928, | |
| "grad_norm": 16.328895540705197, | |
| "learning_rate": 4.979458620695505e-08, | |
| "logits/chosen": -2.029468059539795, | |
| "logits/rejected": -2.0152204036712646, | |
| "logps/chosen": -1.0948221683502197, | |
| "logps/rejected": -1.208194613456726, | |
| "loss": 1.2094, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.1896443367004395, | |
| "rewards/margins": 0.22674505412578583, | |
| "rewards/rejected": -2.416389226913452, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.41786743515850144, | |
| "grad_norm": 19.61140460251683, | |
| "learning_rate": 4.976690039804555e-08, | |
| "logits/chosen": -2.033027172088623, | |
| "logits/rejected": -2.0314948558807373, | |
| "logps/chosen": -0.9877282381057739, | |
| "logps/rejected": -1.0673277378082275, | |
| "loss": 1.2473, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.9754564762115479, | |
| "rewards/margins": 0.1591992825269699, | |
| "rewards/rejected": -2.134655475616455, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.4250720461095101, | |
| "grad_norm": 21.430631009789273, | |
| "learning_rate": 4.973747357830592e-08, | |
| "logits/chosen": -2.0215108394622803, | |
| "logits/rejected": -2.021780490875244, | |
| "logps/chosen": -1.0275431871414185, | |
| "logps/rejected": -1.1647249460220337, | |
| "loss": 1.1677, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -2.055086374282837, | |
| "rewards/margins": 0.2743634283542633, | |
| "rewards/rejected": -2.3294498920440674, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.4322766570605187, | |
| "grad_norm": 19.463998303694815, | |
| "learning_rate": 4.970630781632009e-08, | |
| "logits/chosen": -2.0801994800567627, | |
| "logits/rejected": -2.076254367828369, | |
| "logps/chosen": -1.0327340364456177, | |
| "logps/rejected": -1.1751863956451416, | |
| "loss": 1.1681, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0654680728912354, | |
| "rewards/margins": 0.28490471839904785, | |
| "rewards/rejected": -2.350372791290283, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.43948126801152737, | |
| "grad_norm": 21.00995063503415, | |
| "learning_rate": 4.967340530291242e-08, | |
| "logits/chosen": -2.027909517288208, | |
| "logits/rejected": -2.0180211067199707, | |
| "logps/chosen": -1.0928115844726562, | |
| "logps/rejected": -1.1507136821746826, | |
| "loss": 1.2682, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.1856231689453125, | |
| "rewards/margins": 0.11580429971218109, | |
| "rewards/rejected": -2.3014273643493652, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.44668587896253603, | |
| "grad_norm": 24.905225792062406, | |
| "learning_rate": 4.9638768350993755e-08, | |
| "logits/chosen": -2.0285048484802246, | |
| "logits/rejected": -2.021249771118164, | |
| "logps/chosen": -0.9952943921089172, | |
| "logps/rejected": -1.0829205513000488, | |
| "loss": 1.2345, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -1.9905887842178345, | |
| "rewards/margins": 0.17525213956832886, | |
| "rewards/rejected": -2.1658411026000977, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.4538904899135447, | |
| "grad_norm": 20.771750563160076, | |
| "learning_rate": 4.9602399395398786e-08, | |
| "logits/chosen": -2.0377490520477295, | |
| "logits/rejected": -2.037675380706787, | |
| "logps/chosen": -1.027521014213562, | |
| "logps/rejected": -1.1547839641571045, | |
| "loss": 1.183, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.055042028427124, | |
| "rewards/margins": 0.2545255422592163, | |
| "rewards/rejected": -2.309567928314209, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.4610951008645533, | |
| "grad_norm": 16.17835710154515, | |
| "learning_rate": 4.9564300992714914e-08, | |
| "logits/chosen": -1.9597883224487305, | |
| "logits/rejected": -1.9607963562011719, | |
| "logps/chosen": -1.0108855962753296, | |
| "logps/rejected": -1.116549015045166, | |
| "loss": 1.2101, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.021771192550659, | |
| "rewards/margins": 0.21132683753967285, | |
| "rewards/rejected": -2.233098030090332, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.46829971181556196, | |
| "grad_norm": 21.86769715087536, | |
| "learning_rate": 4.952447582110253e-08, | |
| "logits/chosen": -2.0587735176086426, | |
| "logits/rejected": -2.044377565383911, | |
| "logps/chosen": -1.0383652448654175, | |
| "logps/rejected": -1.1178988218307495, | |
| "loss": 1.2479, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.076730489730835, | |
| "rewards/margins": 0.15906734764575958, | |
| "rewards/rejected": -2.235797643661499, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.4755043227665706, | |
| "grad_norm": 23.755054747254476, | |
| "learning_rate": 4.948292668010676e-08, | |
| "logits/chosen": -2.031721353530884, | |
| "logits/rejected": -2.032727003097534, | |
| "logps/chosen": -1.0880773067474365, | |
| "logps/rejected": -1.1748898029327393, | |
| "loss": 1.2449, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.176154613494873, | |
| "rewards/margins": 0.17362497746944427, | |
| "rewards/rejected": -2.3497796058654785, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.4827089337175792, | |
| "grad_norm": 20.474460354625247, | |
| "learning_rate": 4.943965649046064e-08, | |
| "logits/chosen": -2.0048508644104004, | |
| "logits/rejected": -1.9955081939697266, | |
| "logps/chosen": -1.062713384628296, | |
| "logps/rejected": -1.1663198471069336, | |
| "loss": 1.2154, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.125426769256592, | |
| "rewards/margins": 0.2072126865386963, | |
| "rewards/rejected": -2.332639694213867, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.4899135446685879, | |
| "grad_norm": 19.048186528049722, | |
| "learning_rate": 4.9394668293879835e-08, | |
| "logits/chosen": -1.959315538406372, | |
| "logits/rejected": -1.9503145217895508, | |
| "logps/chosen": -1.0368311405181885, | |
| "logps/rejected": -1.1063206195831299, | |
| "loss": 1.2624, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.073662281036377, | |
| "rewards/margins": 0.13897888362407684, | |
| "rewards/rejected": -2.2126412391662598, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.49711815561959655, | |
| "grad_norm": 24.933354819026505, | |
| "learning_rate": 4.93479652528488e-08, | |
| "logits/chosen": -2.020735502243042, | |
| "logits/rejected": -2.0154590606689453, | |
| "logps/chosen": -1.1052331924438477, | |
| "logps/rejected": -1.209161639213562, | |
| "loss": 1.2262, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.2104663848876953, | |
| "rewards/margins": 0.2078566551208496, | |
| "rewards/rejected": -2.418323278427124, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.5043227665706052, | |
| "grad_norm": 20.317629206968732, | |
| "learning_rate": 4.929955065039848e-08, | |
| "logits/chosen": -2.0213494300842285, | |
| "logits/rejected": -2.0158300399780273, | |
| "logps/chosen": -1.0192697048187256, | |
| "logps/rejected": -1.1514381170272827, | |
| "loss": 1.1829, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.038539409637451, | |
| "rewards/margins": 0.2643369436264038, | |
| "rewards/rejected": -2.3028762340545654, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5115273775216138, | |
| "grad_norm": 19.004922715885144, | |
| "learning_rate": 4.92494278898755e-08, | |
| "logits/chosen": -1.985918402671814, | |
| "logits/rejected": -1.982656717300415, | |
| "logps/chosen": -0.8973722457885742, | |
| "logps/rejected": -1.0216716527938843, | |
| "loss": 1.1973, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -1.7947444915771484, | |
| "rewards/margins": 0.24859857559204102, | |
| "rewards/rejected": -2.0433433055877686, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.5187319884726225, | |
| "grad_norm": 18.960064654240945, | |
| "learning_rate": 4.9197600494702955e-08, | |
| "logits/chosen": -2.007420539855957, | |
| "logits/rejected": -2.001126289367676, | |
| "logps/chosen": -1.0426667928695679, | |
| "logps/rejected": -1.1658456325531006, | |
| "loss": 1.1852, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0853335857391357, | |
| "rewards/margins": 0.2463577687740326, | |
| "rewards/rejected": -2.331691265106201, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.5259365994236311, | |
| "grad_norm": 20.53343043509484, | |
| "learning_rate": 4.9144072108132725e-08, | |
| "logits/chosen": -2.0134854316711426, | |
| "logits/rejected": -2.0023691654205322, | |
| "logps/chosen": -1.0226707458496094, | |
| "logps/rejected": -1.1051828861236572, | |
| "loss": 1.2518, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0453414916992188, | |
| "rewards/margins": 0.16502413153648376, | |
| "rewards/rejected": -2.2103657722473145, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.5331412103746398, | |
| "grad_norm": 17.758862211588106, | |
| "learning_rate": 4.908884649298937e-08, | |
| "logits/chosen": -1.9972114562988281, | |
| "logits/rejected": -2.004119634628296, | |
| "logps/chosen": -1.0192463397979736, | |
| "logps/rejected": -1.0796899795532227, | |
| "loss": 1.2835, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -2.0384926795959473, | |
| "rewards/margins": 0.12088724225759506, | |
| "rewards/rejected": -2.1593799591064453, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.5403458213256485, | |
| "grad_norm": 23.124810759913256, | |
| "learning_rate": 4.903192753140557e-08, | |
| "logits/chosen": -2.0147690773010254, | |
| "logits/rejected": -2.009342908859253, | |
| "logps/chosen": -1.1004369258880615, | |
| "logps/rejected": -1.1906808614730835, | |
| "loss": 1.2378, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -2.200873851776123, | |
| "rewards/margins": 0.18048794567584991, | |
| "rewards/rejected": -2.381361722946167, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.547550432276657, | |
| "grad_norm": 19.72534726379729, | |
| "learning_rate": 4.897331922454931e-08, | |
| "logits/chosen": -1.9795690774917603, | |
| "logits/rejected": -1.9833734035491943, | |
| "logps/chosen": -1.0041850805282593, | |
| "logps/rejected": -1.1136337518692017, | |
| "loss": 1.2165, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0083701610565186, | |
| "rewards/margins": 0.2188970297574997, | |
| "rewards/rejected": -2.2272675037384033, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.5547550432276657, | |
| "grad_norm": 20.539097658978797, | |
| "learning_rate": 4.891302569234256e-08, | |
| "logits/chosen": -1.9727134704589844, | |
| "logits/rejected": -1.9754774570465088, | |
| "logps/chosen": -0.9772794842720032, | |
| "logps/rejected": -1.1290626525878906, | |
| "loss": 1.1643, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -1.9545589685440063, | |
| "rewards/margins": 0.3035663962364197, | |
| "rewards/rejected": -2.2581253051757812, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.5619596541786743, | |
| "grad_norm": 22.07597844396349, | |
| "learning_rate": 4.8851051173171656e-08, | |
| "logits/chosen": -1.9940038919448853, | |
| "logits/rejected": -1.9926246404647827, | |
| "logps/chosen": -1.0405977964401245, | |
| "logps/rejected": -1.1220670938491821, | |
| "loss": 1.2393, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.081195592880249, | |
| "rewards/margins": 0.1629386693239212, | |
| "rewards/rejected": -2.2441341876983643, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.569164265129683, | |
| "grad_norm": 17.470111374688827, | |
| "learning_rate": 4.87874000235894e-08, | |
| "logits/chosen": -2.013667106628418, | |
| "logits/rejected": -2.0078587532043457, | |
| "logps/chosen": -1.0763031244277954, | |
| "logps/rejected": -1.233242154121399, | |
| "loss": 1.1596, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.152606248855591, | |
| "rewards/margins": 0.3138778507709503, | |
| "rewards/rejected": -2.466484308242798, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.5763688760806917, | |
| "grad_norm": 19.520543671943127, | |
| "learning_rate": 4.872207671800876e-08, | |
| "logits/chosen": -2.0354135036468506, | |
| "logits/rejected": -2.0318105220794678, | |
| "logps/chosen": -1.0444949865341187, | |
| "logps/rejected": -1.1220977306365967, | |
| "loss": 1.2567, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0889899730682373, | |
| "rewards/margins": 0.15520496666431427, | |
| "rewards/rejected": -2.2441954612731934, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5835734870317003, | |
| "grad_norm": 15.931565272235597, | |
| "learning_rate": 4.865508584838841e-08, | |
| "logits/chosen": -2.0230934619903564, | |
| "logits/rejected": -2.025510311126709, | |
| "logps/chosen": -1.0136370658874512, | |
| "logps/rejected": -1.1028186082839966, | |
| "loss": 1.2343, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.0272741317749023, | |
| "rewards/margins": 0.17836324870586395, | |
| "rewards/rejected": -2.205637216567993, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.590778097982709, | |
| "grad_norm": 21.101696006896514, | |
| "learning_rate": 4.858643212390985e-08, | |
| "logits/chosen": -2.0232460498809814, | |
| "logits/rejected": -2.0134730339050293, | |
| "logps/chosen": -1.0298725366592407, | |
| "logps/rejected": -1.1151840686798096, | |
| "loss": 1.2504, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.0597450733184814, | |
| "rewards/margins": 0.1706230342388153, | |
| "rewards/rejected": -2.230368137359619, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.5979827089337176, | |
| "grad_norm": 18.384683685983724, | |
| "learning_rate": 4.851612037064643e-08, | |
| "logits/chosen": -2.0008656978607178, | |
| "logits/rejected": -1.9988391399383545, | |
| "logps/chosen": -0.96119225025177, | |
| "logps/rejected": -1.0799301862716675, | |
| "loss": 1.2051, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -1.92238450050354, | |
| "rewards/margins": 0.2374759167432785, | |
| "rewards/rejected": -2.159860372543335, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.6051873198847262, | |
| "grad_norm": 15.976129382373403, | |
| "learning_rate": 4.8444155531224065e-08, | |
| "logits/chosen": -2.0319008827209473, | |
| "logits/rejected": -2.031928539276123, | |
| "logps/chosen": -1.0886750221252441, | |
| "logps/rejected": -1.1605113744735718, | |
| "loss": 1.2625, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.1773500442504883, | |
| "rewards/margins": 0.143672913312912, | |
| "rewards/rejected": -2.3210227489471436, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.6123919308357348, | |
| "grad_norm": 15.372106337343025, | |
| "learning_rate": 4.8370542664473805e-08, | |
| "logits/chosen": -2.03184175491333, | |
| "logits/rejected": -2.0259571075439453, | |
| "logps/chosen": -1.0505023002624512, | |
| "logps/rejected": -1.15494704246521, | |
| "loss": 1.2248, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1010046005249023, | |
| "rewards/margins": 0.20888929069042206, | |
| "rewards/rejected": -2.30989408493042, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.6195965417867435, | |
| "grad_norm": 17.833021138756298, | |
| "learning_rate": 4.829528694507624e-08, | |
| "logits/chosen": -2.011185646057129, | |
| "logits/rejected": -2.0070912837982178, | |
| "logps/chosen": -1.161972999572754, | |
| "logps/rejected": -1.218332290649414, | |
| "loss": 1.28, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.323945999145508, | |
| "rewards/margins": 0.11271880567073822, | |
| "rewards/rejected": -2.436664581298828, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.6268011527377522, | |
| "grad_norm": 20.10043591744987, | |
| "learning_rate": 4.821839366319768e-08, | |
| "logits/chosen": -2.0453944206237793, | |
| "logits/rejected": -2.0392508506774902, | |
| "logps/chosen": -1.0053439140319824, | |
| "logps/rejected": -1.12282395362854, | |
| "loss": 1.1973, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.010687828063965, | |
| "rewards/margins": 0.23495987057685852, | |
| "rewards/rejected": -2.24564790725708, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.6340057636887608, | |
| "grad_norm": 19.633475514009838, | |
| "learning_rate": 4.813986822411833e-08, | |
| "logits/chosen": -2.037318706512451, | |
| "logits/rejected": -2.035334825515747, | |
| "logps/chosen": -1.0152684450149536, | |
| "logps/rejected": -1.0797330141067505, | |
| "loss": 1.2669, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.0305368900299072, | |
| "rewards/margins": 0.12892897427082062, | |
| "rewards/rejected": -2.159466028213501, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.6412103746397695, | |
| "grad_norm": 19.900627573984437, | |
| "learning_rate": 4.805971614785231e-08, | |
| "logits/chosen": -2.0658364295959473, | |
| "logits/rejected": -2.0642929077148438, | |
| "logps/chosen": -1.0170501470565796, | |
| "logps/rejected": -1.11166250705719, | |
| "loss": 1.2213, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.034100294113159, | |
| "rewards/margins": 0.1892244815826416, | |
| "rewards/rejected": -2.22332501411438, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.6484149855907781, | |
| "grad_norm": 20.046745017622534, | |
| "learning_rate": 4.797794306875963e-08, | |
| "logits/chosen": -1.9768317937850952, | |
| "logits/rejected": -1.9782040119171143, | |
| "logps/chosen": -1.1424155235290527, | |
| "logps/rejected": -1.2143452167510986, | |
| "loss": 1.2686, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.2848310470581055, | |
| "rewards/margins": 0.14385904371738434, | |
| "rewards/rejected": -2.4286904335021973, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6556195965417867, | |
| "grad_norm": 20.156486798671747, | |
| "learning_rate": 4.7894554735150076e-08, | |
| "logits/chosen": -1.979318618774414, | |
| "logits/rejected": -1.9829566478729248, | |
| "logps/chosen": -1.042389154434204, | |
| "logps/rejected": -1.108424186706543, | |
| "loss": 1.2626, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.084778308868408, | |
| "rewards/margins": 0.1320703774690628, | |
| "rewards/rejected": -2.216848373413086, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.6628242074927954, | |
| "grad_norm": 23.296556306421977, | |
| "learning_rate": 4.7809557008879185e-08, | |
| "logits/chosen": -2.017183780670166, | |
| "logits/rejected": -2.0119588375091553, | |
| "logps/chosen": -0.9740872383117676, | |
| "logps/rejected": -1.0616848468780518, | |
| "loss": 1.2388, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -1.9481744766235352, | |
| "rewards/margins": 0.1751951277256012, | |
| "rewards/rejected": -2.1233696937561035, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.670028818443804, | |
| "grad_norm": 18.069785801871536, | |
| "learning_rate": 4.772295586493613e-08, | |
| "logits/chosen": -2.057365894317627, | |
| "logits/rejected": -2.054624080657959, | |
| "logps/chosen": -1.0349071025848389, | |
| "logps/rejected": -1.1510379314422607, | |
| "loss": 1.193, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.0698142051696777, | |
| "rewards/margins": 0.23226144909858704, | |
| "rewards/rejected": -2.3020758628845215, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.6772334293948127, | |
| "grad_norm": 19.620026043686646, | |
| "learning_rate": 4.763475739102374e-08, | |
| "logits/chosen": -2.00927472114563, | |
| "logits/rejected": -2.015021562576294, | |
| "logps/chosen": -1.1269561052322388, | |
| "logps/rejected": -1.1944589614868164, | |
| "loss": 1.2561, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.2539122104644775, | |
| "rewards/margins": 0.13500596582889557, | |
| "rewards/rejected": -2.388917922973633, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.6844380403458213, | |
| "grad_norm": 15.430566823053855, | |
| "learning_rate": 4.754496778713054e-08, | |
| "logits/chosen": -1.9693466424942017, | |
| "logits/rejected": -1.9732694625854492, | |
| "logps/chosen": -1.0118048191070557, | |
| "logps/rejected": -1.1344263553619385, | |
| "loss": 1.2008, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0236096382141113, | |
| "rewards/margins": 0.24524304270744324, | |
| "rewards/rejected": -2.268852710723877, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.69164265129683, | |
| "grad_norm": 21.25135809120288, | |
| "learning_rate": 4.7453593365094926e-08, | |
| "logits/chosen": -2.04045033454895, | |
| "logits/rejected": -2.039541244506836, | |
| "logps/chosen": -1.049902319908142, | |
| "logps/rejected": -1.1590924263000488, | |
| "loss": 1.2091, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.099804639816284, | |
| "rewards/margins": 0.21838030219078064, | |
| "rewards/rejected": -2.3181848526000977, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.6988472622478387, | |
| "grad_norm": 21.39072451404026, | |
| "learning_rate": 4.736064054816145e-08, | |
| "logits/chosen": -2.042609691619873, | |
| "logits/rejected": -2.0387399196624756, | |
| "logps/chosen": -0.9685258865356445, | |
| "logps/rejected": -1.0943108797073364, | |
| "loss": 1.1795, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -1.937051773071289, | |
| "rewards/margins": 0.25157004594802856, | |
| "rewards/rejected": -2.188621759414673, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.7060518731988472, | |
| "grad_norm": 17.20168162072602, | |
| "learning_rate": 4.726611587052933e-08, | |
| "logits/chosen": -1.9772526025772095, | |
| "logits/rejected": -1.9768762588500977, | |
| "logps/chosen": -1.1084102392196655, | |
| "logps/rejected": -1.2353932857513428, | |
| "loss": 1.1801, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.216820478439331, | |
| "rewards/margins": 0.2539660334587097, | |
| "rewards/rejected": -2.4707865715026855, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.7132564841498559, | |
| "grad_norm": 22.219628346195623, | |
| "learning_rate": 4.71700259768931e-08, | |
| "logits/chosen": -2.0274641513824463, | |
| "logits/rejected": -2.0244956016540527, | |
| "logps/chosen": -1.109879732131958, | |
| "logps/rejected": -1.2068617343902588, | |
| "loss": 1.2336, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -2.219759464263916, | |
| "rewards/margins": 0.19396351277828217, | |
| "rewards/rejected": -2.4137234687805176, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.7204610951008645, | |
| "grad_norm": 19.81819744621828, | |
| "learning_rate": 4.707237762197549e-08, | |
| "logits/chosen": -2.013184070587158, | |
| "logits/rejected": -2.0100245475769043, | |
| "logps/chosen": -1.0080925226211548, | |
| "logps/rejected": -1.1278679370880127, | |
| "loss": 1.2121, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0161850452423096, | |
| "rewards/margins": 0.23955106735229492, | |
| "rewards/rejected": -2.2557358741760254, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7276657060518732, | |
| "grad_norm": 23.320316952087914, | |
| "learning_rate": 4.697317767005265e-08, | |
| "logits/chosen": -2.0245862007141113, | |
| "logits/rejected": -2.0211946964263916, | |
| "logps/chosen": -1.002010703086853, | |
| "logps/rejected": -1.0942790508270264, | |
| "loss": 1.2568, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -2.004021406173706, | |
| "rewards/margins": 0.1845366507768631, | |
| "rewards/rejected": -2.1885581016540527, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.7348703170028819, | |
| "grad_norm": 17.35614684932965, | |
| "learning_rate": 4.6872433094471577e-08, | |
| "logits/chosen": -2.0214576721191406, | |
| "logits/rejected": -2.01664137840271, | |
| "logps/chosen": -1.0324314832687378, | |
| "logps/rejected": -1.127612590789795, | |
| "loss": 1.2122, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0648629665374756, | |
| "rewards/margins": 0.19036227464675903, | |
| "rewards/rejected": -2.25522518157959, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.7420749279538905, | |
| "grad_norm": 16.487356163413914, | |
| "learning_rate": 4.677015097715994e-08, | |
| "logits/chosen": -1.9668807983398438, | |
| "logits/rejected": -1.9662902355194092, | |
| "logps/chosen": -1.0229971408843994, | |
| "logps/rejected": -1.1552445888519287, | |
| "loss": 1.1997, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -2.045994281768799, | |
| "rewards/margins": 0.2644946873188019, | |
| "rewards/rejected": -2.3104891777038574, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.7492795389048992, | |
| "grad_norm": 17.492033929105126, | |
| "learning_rate": 4.666633850812825e-08, | |
| "logits/chosen": -2.0216238498687744, | |
| "logits/rejected": -2.0157718658447266, | |
| "logps/chosen": -1.0129607915878296, | |
| "logps/rejected": -1.0947651863098145, | |
| "loss": 1.2367, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.025921583175659, | |
| "rewards/margins": 0.16360855102539062, | |
| "rewards/rejected": -2.189530372619629, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.7564841498559077, | |
| "grad_norm": 17.49180259130834, | |
| "learning_rate": 4.656100298496439e-08, | |
| "logits/chosen": -1.971518874168396, | |
| "logits/rejected": -1.9679629802703857, | |
| "logps/chosen": -0.9385242462158203, | |
| "logps/rejected": -1.0688835382461548, | |
| "loss": 1.1859, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.8770484924316406, | |
| "rewards/margins": 0.26071876287460327, | |
| "rewards/rejected": -2.1377670764923096, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.7636887608069164, | |
| "grad_norm": 17.905832545876255, | |
| "learning_rate": 4.6454151812320715e-08, | |
| "logits/chosen": -2.0001180171966553, | |
| "logits/rejected": -1.9940083026885986, | |
| "logps/chosen": -1.03890061378479, | |
| "logps/rejected": -1.1473093032836914, | |
| "loss": 1.2178, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.07780122756958, | |
| "rewards/margins": 0.21681778132915497, | |
| "rewards/rejected": -2.294618606567383, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.770893371757925, | |
| "grad_norm": 20.829212072329433, | |
| "learning_rate": 4.6345792501393434e-08, | |
| "logits/chosen": -2.0026588439941406, | |
| "logits/rejected": -2.0007362365722656, | |
| "logps/chosen": -1.0745230913162231, | |
| "logps/rejected": -1.201542615890503, | |
| "loss": 1.2046, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.1490461826324463, | |
| "rewards/margins": 0.2540392279624939, | |
| "rewards/rejected": -2.403085231781006, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.7780979827089337, | |
| "grad_norm": 20.734671350383845, | |
| "learning_rate": 4.6235932669394676e-08, | |
| "logits/chosen": -2.0293848514556885, | |
| "logits/rejected": -2.030176877975464, | |
| "logps/chosen": -1.0878403186798096, | |
| "logps/rejected": -1.196656584739685, | |
| "loss": 1.2186, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.175680637359619, | |
| "rewards/margins": 0.21763241291046143, | |
| "rewards/rejected": -2.39331316947937, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.7853025936599424, | |
| "grad_norm": 24.317214064629283, | |
| "learning_rate": 4.612458003901698e-08, | |
| "logits/chosen": -2.041074514389038, | |
| "logits/rejected": -2.0332765579223633, | |
| "logps/chosen": -1.109058141708374, | |
| "logps/rejected": -1.2108246088027954, | |
| "loss": 1.2286, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.218116283416748, | |
| "rewards/margins": 0.20353302359580994, | |
| "rewards/rejected": -2.421649217605591, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.792507204610951, | |
| "grad_norm": 23.34024566936978, | |
| "learning_rate": 4.6011742437890476e-08, | |
| "logits/chosen": -2.028428077697754, | |
| "logits/rejected": -2.023019790649414, | |
| "logps/chosen": -1.0458049774169922, | |
| "logps/rejected": -1.1794006824493408, | |
| "loss": 1.1775, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0916099548339844, | |
| "rewards/margins": 0.2671913504600525, | |
| "rewards/rejected": -2.3588013648986816, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7997118155619597, | |
| "grad_norm": 16.933139927466357, | |
| "learning_rate": 4.589742779803259e-08, | |
| "logits/chosen": -2.025526523590088, | |
| "logits/rejected": -2.018397569656372, | |
| "logps/chosen": -1.0093412399291992, | |
| "logps/rejected": -1.129741907119751, | |
| "loss": 1.1948, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0186824798583984, | |
| "rewards/margins": 0.2408013790845871, | |
| "rewards/rejected": -2.259483814239502, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.8069164265129684, | |
| "grad_norm": 18.433386982266423, | |
| "learning_rate": 4.5781644155290486e-08, | |
| "logits/chosen": -1.9837512969970703, | |
| "logits/rejected": -1.9759635925292969, | |
| "logps/chosen": -1.047893762588501, | |
| "logps/rejected": -1.1082303524017334, | |
| "loss": 1.2713, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.095787525177002, | |
| "rewards/margins": 0.12067310512065887, | |
| "rewards/rejected": -2.216460704803467, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.8141210374639769, | |
| "grad_norm": 18.152544924178944, | |
| "learning_rate": 4.566439964877613e-08, | |
| "logits/chosen": -2.0132524967193604, | |
| "logits/rejected": -2.0092389583587646, | |
| "logps/chosen": -0.9992140531539917, | |
| "logps/rejected": -1.0850255489349365, | |
| "loss": 1.2443, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.9984281063079834, | |
| "rewards/margins": 0.17162318527698517, | |
| "rewards/rejected": -2.170051097869873, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.8213256484149856, | |
| "grad_norm": 16.195560643437258, | |
| "learning_rate": 4.554570252029421e-08, | |
| "logits/chosen": -2.0523180961608887, | |
| "logits/rejected": -2.0510833263397217, | |
| "logps/chosen": -1.0488303899765015, | |
| "logps/rejected": -1.1647334098815918, | |
| "loss": 1.201, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.097660779953003, | |
| "rewards/margins": 0.23180584609508514, | |
| "rewards/rejected": -2.3294668197631836, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.8285302593659942, | |
| "grad_norm": 17.9745846350065, | |
| "learning_rate": 4.542556111376274e-08, | |
| "logits/chosen": -2.0492236614227295, | |
| "logits/rejected": -2.0428290367126465, | |
| "logps/chosen": -1.0749974250793457, | |
| "logps/rejected": -1.166634202003479, | |
| "loss": 1.2395, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.1499948501586914, | |
| "rewards/margins": 0.18327349424362183, | |
| "rewards/rejected": -2.333268404006958, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.8357348703170029, | |
| "grad_norm": 23.066926614034124, | |
| "learning_rate": 4.5303983874626506e-08, | |
| "logits/chosen": -1.9916588068008423, | |
| "logits/rejected": -1.9900974035263062, | |
| "logps/chosen": -1.0387059450149536, | |
| "logps/rejected": -1.115934133529663, | |
| "loss": 1.2651, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.0774118900299072, | |
| "rewards/margins": 0.15445652604103088, | |
| "rewards/rejected": -2.231868267059326, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.8429394812680115, | |
| "grad_norm": 20.106291828506194, | |
| "learning_rate": 4.518097934926339e-08, | |
| "logits/chosen": -1.995008111000061, | |
| "logits/rejected": -1.9863529205322266, | |
| "logps/chosen": -1.0165393352508545, | |
| "logps/rejected": -1.1260240077972412, | |
| "loss": 1.2057, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.033078670501709, | |
| "rewards/margins": 0.21896927058696747, | |
| "rewards/rejected": -2.2520480155944824, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.8501440922190202, | |
| "grad_norm": 22.8857527390999, | |
| "learning_rate": 4.505655618438363e-08, | |
| "logits/chosen": -1.9628753662109375, | |
| "logits/rejected": -1.9588840007781982, | |
| "logps/chosen": -1.0615794658660889, | |
| "logps/rejected": -1.1646844148635864, | |
| "loss": 1.2307, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.1231589317321777, | |
| "rewards/margins": 0.20620973408222198, | |
| "rewards/rejected": -2.329368829727173, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.8573487031700289, | |
| "grad_norm": 17.434899766590377, | |
| "learning_rate": 4.4930723126421945e-08, | |
| "logits/chosen": -2.0546653270721436, | |
| "logits/rejected": -2.047938585281372, | |
| "logps/chosen": -1.0720479488372803, | |
| "logps/rejected": -1.1471359729766846, | |
| "loss": 1.252, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.1440958976745605, | |
| "rewards/margins": 0.15017575025558472, | |
| "rewards/rejected": -2.294271945953369, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.8645533141210374, | |
| "grad_norm": 22.14075015263452, | |
| "learning_rate": 4.48034890209227e-08, | |
| "logits/chosen": -1.983888030052185, | |
| "logits/rejected": -1.9716113805770874, | |
| "logps/chosen": -1.0871121883392334, | |
| "logps/rejected": -1.1737545728683472, | |
| "loss": 1.23, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.174224376678467, | |
| "rewards/margins": 0.1732848584651947, | |
| "rewards/rejected": -2.3475091457366943, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.8717579250720461, | |
| "grad_norm": 18.53077050982448, | |
| "learning_rate": 4.4674862811918155e-08, | |
| "logits/chosen": -1.971573829650879, | |
| "logits/rejected": -1.980055570602417, | |
| "logps/chosen": -0.9388012886047363, | |
| "logps/rejected": -1.091797113418579, | |
| "loss": 1.1596, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.8776025772094727, | |
| "rewards/margins": 0.30599164962768555, | |
| "rewards/rejected": -2.183594226837158, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.8789625360230547, | |
| "grad_norm": 17.441252552193376, | |
| "learning_rate": 4.454485354129966e-08, | |
| "logits/chosen": -1.9985713958740234, | |
| "logits/rejected": -1.994210958480835, | |
| "logps/chosen": -1.0104329586029053, | |
| "logps/rejected": -1.11543869972229, | |
| "loss": 1.2194, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.0208659172058105, | |
| "rewards/margins": 0.21001139283180237, | |
| "rewards/rejected": -2.23087739944458, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.8861671469740634, | |
| "grad_norm": 17.20275284474546, | |
| "learning_rate": 4.4413470348182124e-08, | |
| "logits/chosen": -1.9755537509918213, | |
| "logits/rejected": -1.9634100198745728, | |
| "logps/chosen": -0.9853811264038086, | |
| "logps/rejected": -1.076774001121521, | |
| "loss": 1.2316, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.9707622528076172, | |
| "rewards/margins": 0.18278571963310242, | |
| "rewards/rejected": -2.153548002243042, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.8933717579250721, | |
| "grad_norm": 21.253905408711432, | |
| "learning_rate": 4.42807224682615e-08, | |
| "logits/chosen": -1.9815731048583984, | |
| "logits/rejected": -1.9793331623077393, | |
| "logps/chosen": -0.9373159408569336, | |
| "logps/rejected": -1.0729162693023682, | |
| "loss": 1.181, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.8746318817138672, | |
| "rewards/margins": 0.2712007462978363, | |
| "rewards/rejected": -2.1458325386047363, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.9005763688760807, | |
| "grad_norm": 18.803145183231678, | |
| "learning_rate": 4.4146619233165604e-08, | |
| "logits/chosen": -2.0230329036712646, | |
| "logits/rejected": -2.025296688079834, | |
| "logps/chosen": -1.0652821063995361, | |
| "logps/rejected": -1.2190508842468262, | |
| "loss": 1.1677, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.1305642127990723, | |
| "rewards/margins": 0.3075374960899353, | |
| "rewards/rejected": -2.4381017684936523, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.9077809798270894, | |
| "grad_norm": 25.018490567837954, | |
| "learning_rate": 4.4011170069798126e-08, | |
| "logits/chosen": -2.020940065383911, | |
| "logits/rejected": -2.025850296020508, | |
| "logps/chosen": -1.1181256771087646, | |
| "logps/rejected": -1.2433640956878662, | |
| "loss": 1.1932, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.2362513542175293, | |
| "rewards/margins": 0.25047701597213745, | |
| "rewards/rejected": -2.4867281913757324, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.9149855907780979, | |
| "grad_norm": 18.123087760553187, | |
| "learning_rate": 4.387438449967594e-08, | |
| "logits/chosen": -1.982254981994629, | |
| "logits/rejected": -1.97560715675354, | |
| "logps/chosen": -0.9658223986625671, | |
| "logps/rejected": -1.085925579071045, | |
| "loss": 1.1909, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -1.9316447973251343, | |
| "rewards/margins": 0.24020643532276154, | |
| "rewards/rejected": -2.17185115814209, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.9221902017291066, | |
| "grad_norm": 21.17056826903978, | |
| "learning_rate": 4.373627213825983e-08, | |
| "logits/chosen": -2.0719313621520996, | |
| "logits/rejected": -2.0676798820495605, | |
| "logps/chosen": -1.0272830724716187, | |
| "logps/rejected": -1.1627672910690308, | |
| "loss": 1.1829, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0545661449432373, | |
| "rewards/margins": 0.27096837759017944, | |
| "rewards/rejected": -2.3255345821380615, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.9293948126801153, | |
| "grad_norm": 16.73743221772608, | |
| "learning_rate": 4.359684269427848e-08, | |
| "logits/chosen": -2.034970760345459, | |
| "logits/rejected": -2.0339713096618652, | |
| "logps/chosen": -0.9956309199333191, | |
| "logps/rejected": -1.0993244647979736, | |
| "loss": 1.2107, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.9912618398666382, | |
| "rewards/margins": 0.20738673210144043, | |
| "rewards/rejected": -2.1986489295959473, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.9365994236311239, | |
| "grad_norm": 23.479698749807888, | |
| "learning_rate": 4.34561059690461e-08, | |
| "logits/chosen": -2.079378843307495, | |
| "logits/rejected": -2.0813305377960205, | |
| "logps/chosen": -1.047837495803833, | |
| "logps/rejected": -1.112128496170044, | |
| "loss": 1.2707, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -2.095674991607666, | |
| "rewards/margins": 0.12858203053474426, | |
| "rewards/rejected": -2.224256992340088, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.9438040345821326, | |
| "grad_norm": 21.373476828454745, | |
| "learning_rate": 4.3314071855773314e-08, | |
| "logits/chosen": -2.044544219970703, | |
| "logits/rejected": -2.0450897216796875, | |
| "logps/chosen": -0.9845747947692871, | |
| "logps/rejected": -1.0792670249938965, | |
| "loss": 1.2235, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -1.9691495895385742, | |
| "rewards/margins": 0.18938450515270233, | |
| "rewards/rejected": -2.158534049987793, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.9510086455331412, | |
| "grad_norm": 20.390675123621403, | |
| "learning_rate": 4.3170750338871806e-08, | |
| "logits/chosen": -2.0153450965881348, | |
| "logits/rejected": -2.008953094482422, | |
| "logps/chosen": -1.0770236253738403, | |
| "logps/rejected": -1.2197729349136353, | |
| "loss": 1.1662, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.1540472507476807, | |
| "rewards/margins": 0.2854984402656555, | |
| "rewards/rejected": -2.4395458698272705, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.9582132564841499, | |
| "grad_norm": 14.835531781677203, | |
| "learning_rate": 4.3026151493252414e-08, | |
| "logits/chosen": -2.039367437362671, | |
| "logits/rejected": -2.0349154472351074, | |
| "logps/chosen": -1.0609954595565796, | |
| "logps/rejected": -1.1818583011627197, | |
| "loss": 1.2003, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.121990919113159, | |
| "rewards/margins": 0.24172568321228027, | |
| "rewards/rejected": -2.3637166023254395, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.9654178674351584, | |
| "grad_norm": 25.71038185604989, | |
| "learning_rate": 4.2880285483616895e-08, | |
| "logits/chosen": -2.0069704055786133, | |
| "logits/rejected": -2.007664680480957, | |
| "logps/chosen": -1.0175052881240845, | |
| "logps/rejected": -1.1325770616531372, | |
| "loss": 1.2093, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.035010576248169, | |
| "rewards/margins": 0.2301437109708786, | |
| "rewards/rejected": -2.2651541233062744, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.9726224783861671, | |
| "grad_norm": 16.092608904878997, | |
| "learning_rate": 4.273316256374342e-08, | |
| "logits/chosen": -1.9464366436004639, | |
| "logits/rejected": -1.9446899890899658, | |
| "logps/chosen": -1.01396644115448, | |
| "logps/rejected": -1.0869011878967285, | |
| "loss": 1.264, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.02793288230896, | |
| "rewards/margins": 0.1458693891763687, | |
| "rewards/rejected": -2.173802375793457, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.9798270893371758, | |
| "grad_norm": 16.212857235886922, | |
| "learning_rate": 4.258479307576576e-08, | |
| "logits/chosen": -1.9840402603149414, | |
| "logits/rejected": -1.9818894863128662, | |
| "logps/chosen": -0.9638694524765015, | |
| "logps/rejected": -1.0546468496322632, | |
| "loss": 1.24, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -1.927738904953003, | |
| "rewards/margins": 0.18155473470687866, | |
| "rewards/rejected": -2.1092936992645264, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.9870317002881844, | |
| "grad_norm": 21.341000872382455, | |
| "learning_rate": 4.243518744944626e-08, | |
| "logits/chosen": -2.0093555450439453, | |
| "logits/rejected": -2.0047600269317627, | |
| "logps/chosen": -1.0009874105453491, | |
| "logps/rejected": -1.1216598749160767, | |
| "loss": 1.1889, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0019748210906982, | |
| "rewards/margins": 0.24134452641010284, | |
| "rewards/rejected": -2.2433197498321533, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.9942363112391931, | |
| "grad_norm": 20.83793747644969, | |
| "learning_rate": 4.22843562014427e-08, | |
| "logits/chosen": -1.9709367752075195, | |
| "logits/rejected": -1.9672348499298096, | |
| "logps/chosen": -1.0514830350875854, | |
| "logps/rejected": -1.1256954669952393, | |
| "loss": 1.2503, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.102966070175171, | |
| "rewards/margins": 0.14842486381530762, | |
| "rewards/rejected": -2.2513909339904785, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.0014409221902016, | |
| "grad_norm": 28.539886189287515, | |
| "learning_rate": 4.2132309934569e-08, | |
| "logits/chosen": -2.051409959793091, | |
| "logits/rejected": -2.051856517791748, | |
| "logps/chosen": -1.015867829322815, | |
| "logps/rejected": -1.128615379333496, | |
| "loss": 1.2103, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.03173565864563, | |
| "rewards/margins": 0.22549493610858917, | |
| "rewards/rejected": -2.257230758666992, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.0086455331412103, | |
| "grad_norm": 18.4548159325349, | |
| "learning_rate": 4.197905933704989e-08, | |
| "logits/chosen": -1.9460541009902954, | |
| "logits/rejected": -1.9434579610824585, | |
| "logps/chosen": -1.0608787536621094, | |
| "logps/rejected": -1.1942651271820068, | |
| "loss": 1.2017, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1217575073242188, | |
| "rewards/margins": 0.2667728066444397, | |
| "rewards/rejected": -2.3885302543640137, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.015850144092219, | |
| "grad_norm": 23.677146712392545, | |
| "learning_rate": 4.1824615181769577e-08, | |
| "logits/chosen": -1.992706060409546, | |
| "logits/rejected": -1.9971202611923218, | |
| "logps/chosen": -1.0128545761108398, | |
| "logps/rejected": -1.138115644454956, | |
| "loss": 1.2035, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0257091522216797, | |
| "rewards/margins": 0.2505221366882324, | |
| "rewards/rejected": -2.276231288909912, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.0230547550432276, | |
| "grad_norm": 18.466913113268376, | |
| "learning_rate": 4.1668988325514434e-08, | |
| "logits/chosen": -2.0149149894714355, | |
| "logits/rejected": -2.0098109245300293, | |
| "logps/chosen": -1.1170905828475952, | |
| "logps/rejected": -1.2321112155914307, | |
| "loss": 1.2252, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -2.2341811656951904, | |
| "rewards/margins": 0.2300410270690918, | |
| "rewards/rejected": -2.4642224311828613, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.0302593659942363, | |
| "grad_norm": 20.916480925982736, | |
| "learning_rate": 4.1512189708209844e-08, | |
| "logits/chosen": -2.0576863288879395, | |
| "logits/rejected": -2.0563559532165527, | |
| "logps/chosen": -0.9412269592285156, | |
| "logps/rejected": -1.0276962518692017, | |
| "loss": 1.2464, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -1.8824539184570312, | |
| "rewards/margins": 0.17293845117092133, | |
| "rewards/rejected": -2.0553925037384033, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.037463976945245, | |
| "grad_norm": 22.10230375057076, | |
| "learning_rate": 4.1354230352151143e-08, | |
| "logits/chosen": -2.009265661239624, | |
| "logits/rejected": -2.002540111541748, | |
| "logps/chosen": -1.138351559638977, | |
| "logps/rejected": -1.2199509143829346, | |
| "loss": 1.2585, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.276703119277954, | |
| "rewards/margins": 0.16319862008094788, | |
| "rewards/rejected": -2.439901828765869, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.0446685878962536, | |
| "grad_norm": 16.842031017248782, | |
| "learning_rate": 4.119512136122882e-08, | |
| "logits/chosen": -2.0774807929992676, | |
| "logits/rejected": -2.086643695831299, | |
| "logps/chosen": -0.9951038360595703, | |
| "logps/rejected": -1.1451139450073242, | |
| "loss": 1.1708, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9902076721191406, | |
| "rewards/margins": 0.30002015829086304, | |
| "rewards/rejected": -2.2902278900146484, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.0518731988472623, | |
| "grad_norm": 15.427164808054908, | |
| "learning_rate": 4.103487392014795e-08, | |
| "logits/chosen": -1.992767095565796, | |
| "logits/rejected": -1.980544090270996, | |
| "logps/chosen": -1.0006814002990723, | |
| "logps/rejected": -1.15886390209198, | |
| "loss": 1.1455, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.0013628005981445, | |
| "rewards/margins": 0.31636515259742737, | |
| "rewards/rejected": -2.31772780418396, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.059077809798271, | |
| "grad_norm": 16.81042888795935, | |
| "learning_rate": 4.087349929364192e-08, | |
| "logits/chosen": -2.034682273864746, | |
| "logits/rejected": -2.0252864360809326, | |
| "logps/chosen": -0.9601753354072571, | |
| "logps/rejected": -1.0913857221603394, | |
| "loss": 1.1863, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -1.9203506708145142, | |
| "rewards/margins": 0.26242080330848694, | |
| "rewards/rejected": -2.1827714443206787, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.0662824207492796, | |
| "grad_norm": 17.584619579081235, | |
| "learning_rate": 4.0711008825680645e-08, | |
| "logits/chosen": -1.979069709777832, | |
| "logits/rejected": -1.97795832157135, | |
| "logps/chosen": -1.0063145160675049, | |
| "logps/rejected": -1.1248080730438232, | |
| "loss": 1.2064, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0126290321350098, | |
| "rewards/margins": 0.2369869500398636, | |
| "rewards/rejected": -2.2496161460876465, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.0734870317002883, | |
| "grad_norm": 19.98068478862068, | |
| "learning_rate": 4.054741393867306e-08, | |
| "logits/chosen": -1.99558424949646, | |
| "logits/rejected": -1.9926925897598267, | |
| "logps/chosen": -1.1117796897888184, | |
| "logps/rejected": -1.1623036861419678, | |
| "loss": 1.2882, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.2235593795776367, | |
| "rewards/margins": 0.10104763507843018, | |
| "rewards/rejected": -2.3246073722839355, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.080691642651297, | |
| "grad_norm": 18.569188294062595, | |
| "learning_rate": 4.038272613266419e-08, | |
| "logits/chosen": -1.9959461688995361, | |
| "logits/rejected": -1.9826500415802002, | |
| "logps/chosen": -1.0095350742340088, | |
| "logps/rejected": -1.1202278137207031, | |
| "loss": 1.2023, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.0190701484680176, | |
| "rewards/margins": 0.22138550877571106, | |
| "rewards/rejected": -2.2404556274414062, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0878962536023056, | |
| "grad_norm": 18.129783454014866, | |
| "learning_rate": 4.0216956984526784e-08, | |
| "logits/chosen": -2.04606032371521, | |
| "logits/rejected": -2.047947406768799, | |
| "logps/chosen": -1.0161449909210205, | |
| "logps/rejected": -1.124267339706421, | |
| "loss": 1.2167, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.032289981842041, | |
| "rewards/margins": 0.21624493598937988, | |
| "rewards/rejected": -2.248534679412842, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.0951008645533142, | |
| "grad_norm": 16.171374987629033, | |
| "learning_rate": 4.0050118147147446e-08, | |
| "logits/chosen": -1.9890464544296265, | |
| "logits/rejected": -1.989335298538208, | |
| "logps/chosen": -1.0982977151870728, | |
| "logps/rejected": -1.110621690750122, | |
| "loss": 1.3393, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -2.1965954303741455, | |
| "rewards/margins": 0.02464829757809639, | |
| "rewards/rejected": -2.221243381500244, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.1023054755043227, | |
| "grad_norm": 17.66132069219183, | |
| "learning_rate": 3.988222134860755e-08, | |
| "logits/chosen": -2.0323548316955566, | |
| "logits/rejected": -2.0236430168151855, | |
| "logps/chosen": -0.9508152008056641, | |
| "logps/rejected": -1.1158647537231445, | |
| "loss": 1.1407, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.9016304016113281, | |
| "rewards/margins": 0.33009934425354004, | |
| "rewards/rejected": -2.231729507446289, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.1095100864553313, | |
| "grad_norm": 23.676130358664636, | |
| "learning_rate": 3.9713278391358724e-08, | |
| "logits/chosen": -2.0360183715820312, | |
| "logits/rejected": -2.0298333168029785, | |
| "logps/chosen": -1.025137186050415, | |
| "logps/rejected": -1.1484403610229492, | |
| "loss": 1.1877, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.05027437210083, | |
| "rewards/margins": 0.24660632014274597, | |
| "rewards/rejected": -2.2968807220458984, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.11671469740634, | |
| "grad_norm": 17.77840056029204, | |
| "learning_rate": 3.954330115139328e-08, | |
| "logits/chosen": -2.0122570991516113, | |
| "logits/rejected": -2.0070974826812744, | |
| "logps/chosen": -1.0277431011199951, | |
| "logps/rejected": -1.1330978870391846, | |
| "loss": 1.2216, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0554862022399902, | |
| "rewards/margins": 0.21070995926856995, | |
| "rewards/rejected": -2.266195774078369, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.1239193083573487, | |
| "grad_norm": 25.812098081681867, | |
| "learning_rate": 3.937230157740931e-08, | |
| "logits/chosen": -2.067347764968872, | |
| "logits/rejected": -2.0611376762390137, | |
| "logps/chosen": -1.0478734970092773, | |
| "logps/rejected": -1.1832859516143799, | |
| "loss": 1.1824, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0957469940185547, | |
| "rewards/margins": 0.2708250880241394, | |
| "rewards/rejected": -2.3665719032287598, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.1311239193083573, | |
| "grad_norm": 16.22328310375803, | |
| "learning_rate": 3.920029168997077e-08, | |
| "logits/chosen": -2.04835844039917, | |
| "logits/rejected": -2.04648494720459, | |
| "logps/chosen": -1.0037837028503418, | |
| "logps/rejected": -1.131502628326416, | |
| "loss": 1.1863, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.0075674057006836, | |
| "rewards/margins": 0.25543779134750366, | |
| "rewards/rejected": -2.263005256652832, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 1.138328530259366, | |
| "grad_norm": 29.81353401958458, | |
| "learning_rate": 3.9027283580662476e-08, | |
| "logits/chosen": -2.0225307941436768, | |
| "logits/rejected": -2.0166521072387695, | |
| "logps/chosen": -1.0478241443634033, | |
| "logps/rejected": -1.1936235427856445, | |
| "loss": 1.1765, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.0956482887268066, | |
| "rewards/margins": 0.2915985882282257, | |
| "rewards/rejected": -2.387247085571289, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 1.1455331412103746, | |
| "grad_norm": 16.941588748106863, | |
| "learning_rate": 3.885328941124014e-08, | |
| "logits/chosen": -1.991965889930725, | |
| "logits/rejected": -1.9873225688934326, | |
| "logps/chosen": -0.9666957855224609, | |
| "logps/rejected": -1.1006277799606323, | |
| "loss": 1.1706, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -1.9333915710449219, | |
| "rewards/margins": 0.26786428689956665, | |
| "rewards/rejected": -2.2012555599212646, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 1.1527377521613833, | |
| "grad_norm": 20.82364621838478, | |
| "learning_rate": 3.867832141277539e-08, | |
| "logits/chosen": -2.0321202278137207, | |
| "logits/rejected": -2.0232601165771484, | |
| "logps/chosen": -1.0682156085968018, | |
| "logps/rejected": -1.180410623550415, | |
| "loss": 1.2096, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.1364312171936035, | |
| "rewards/margins": 0.2243901491165161, | |
| "rewards/rejected": -2.36082124710083, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.159942363112392, | |
| "grad_norm": 20.912686096120964, | |
| "learning_rate": 3.850239188479606e-08, | |
| "logits/chosen": -1.9847033023834229, | |
| "logits/rejected": -1.9881378412246704, | |
| "logps/chosen": -1.0096313953399658, | |
| "logps/rejected": -1.1002733707427979, | |
| "loss": 1.2372, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0192627906799316, | |
| "rewards/margins": 0.18128342926502228, | |
| "rewards/rejected": -2.2005467414855957, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 1.1671469740634006, | |
| "grad_norm": 21.899733424702635, | |
| "learning_rate": 3.832551319442151e-08, | |
| "logits/chosen": -2.0586349964141846, | |
| "logits/rejected": -2.059906482696533, | |
| "logps/chosen": -1.057755708694458, | |
| "logps/rejected": -1.184890627861023, | |
| "loss": 1.1897, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.115511417388916, | |
| "rewards/margins": 0.254270076751709, | |
| "rewards/rejected": -2.369781255722046, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.1743515850144093, | |
| "grad_norm": 17.2910410178799, | |
| "learning_rate": 3.81476977754933e-08, | |
| "logits/chosen": -1.9560763835906982, | |
| "logits/rejected": -1.9524368047714233, | |
| "logps/chosen": -1.0269958972930908, | |
| "logps/rejected": -1.0969812870025635, | |
| "loss": 1.2579, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0539917945861816, | |
| "rewards/margins": 0.13997015357017517, | |
| "rewards/rejected": -2.193962574005127, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 1.181556195965418, | |
| "grad_norm": 16.85537517324203, | |
| "learning_rate": 3.796895812770114e-08, | |
| "logits/chosen": -1.9784526824951172, | |
| "logits/rejected": -1.9793262481689453, | |
| "logps/chosen": -1.0173704624176025, | |
| "logps/rejected": -1.1091585159301758, | |
| "loss": 1.2408, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.034740924835205, | |
| "rewards/margins": 0.1835760474205017, | |
| "rewards/rejected": -2.2183170318603516, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 1.1887608069164266, | |
| "grad_norm": 22.175531020521074, | |
| "learning_rate": 3.7789306815704216e-08, | |
| "logits/chosen": -2.009108781814575, | |
| "logits/rejected": -2.006824254989624, | |
| "logps/chosen": -1.0072455406188965, | |
| "logps/rejected": -1.0781667232513428, | |
| "loss": 1.2618, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.014491081237793, | |
| "rewards/margins": 0.14184223115444183, | |
| "rewards/rejected": -2.1563334465026855, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.195965417867435, | |
| "grad_norm": 18.89404553225258, | |
| "learning_rate": 3.760875646824795e-08, | |
| "logits/chosen": -1.932428002357483, | |
| "logits/rejected": -1.9363291263580322, | |
| "logps/chosen": -0.9747514724731445, | |
| "logps/rejected": -1.0793020725250244, | |
| "loss": 1.2232, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.949502944946289, | |
| "rewards/margins": 0.20910124480724335, | |
| "rewards/rejected": -2.158604145050049, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 1.2031700288184437, | |
| "grad_norm": 22.227342153467788, | |
| "learning_rate": 3.742731977727623e-08, | |
| "logits/chosen": -2.0331270694732666, | |
| "logits/rejected": -2.0301709175109863, | |
| "logps/chosen": -1.0413671731948853, | |
| "logps/rejected": -1.1772552728652954, | |
| "loss": 1.1795, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0827343463897705, | |
| "rewards/margins": 0.27177631855010986, | |
| "rewards/rejected": -2.354510545730591, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 1.2103746397694524, | |
| "grad_norm": 19.70333261721218, | |
| "learning_rate": 3.7245009497039244e-08, | |
| "logits/chosen": -1.970715880393982, | |
| "logits/rejected": -1.9627761840820312, | |
| "logps/chosen": -1.0116103887557983, | |
| "logps/rejected": -1.1484659910202026, | |
| "loss": 1.1716, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.0232207775115967, | |
| "rewards/margins": 0.2737112045288086, | |
| "rewards/rejected": -2.2969319820404053, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 1.217579250720461, | |
| "grad_norm": 18.86128397711634, | |
| "learning_rate": 3.7061838443196886e-08, | |
| "logits/chosen": -2.0151665210723877, | |
| "logits/rejected": -2.016679048538208, | |
| "logps/chosen": -1.026761531829834, | |
| "logps/rejected": -1.150320053100586, | |
| "loss": 1.1888, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.053523063659668, | |
| "rewards/margins": 0.2471170723438263, | |
| "rewards/rejected": -2.300640106201172, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 1.2247838616714697, | |
| "grad_norm": 22.70930817597516, | |
| "learning_rate": 3.68778194919179e-08, | |
| "logits/chosen": -1.983304738998413, | |
| "logits/rejected": -1.984287977218628, | |
| "logps/chosen": -1.0792837142944336, | |
| "logps/rejected": -1.2009527683258057, | |
| "loss": 1.1955, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.158567428588867, | |
| "rewards/margins": 0.24333825707435608, | |
| "rewards/rejected": -2.4019055366516113, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.2319884726224783, | |
| "grad_norm": 20.280682845222326, | |
| "learning_rate": 3.66929655789747e-08, | |
| "logits/chosen": -2.0337467193603516, | |
| "logits/rejected": -2.0225701332092285, | |
| "logps/chosen": -0.9402590990066528, | |
| "logps/rejected": -1.0919773578643799, | |
| "loss": 1.1634, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.8805181980133057, | |
| "rewards/margins": 0.30343663692474365, | |
| "rewards/rejected": -2.1839547157287598, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 1.239193083573487, | |
| "grad_norm": 16.359613747281564, | |
| "learning_rate": 3.6507289698834064e-08, | |
| "logits/chosen": -1.9774224758148193, | |
| "logits/rejected": -1.973842978477478, | |
| "logps/chosen": -0.98408442735672, | |
| "logps/rejected": -1.1161837577819824, | |
| "loss": 1.196, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.96816885471344, | |
| "rewards/margins": 0.2641984820365906, | |
| "rewards/rejected": -2.232367515563965, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 1.2463976945244957, | |
| "grad_norm": 25.191044914408238, | |
| "learning_rate": 3.6320804903743684e-08, | |
| "logits/chosen": -2.026642322540283, | |
| "logits/rejected": -2.0262598991394043, | |
| "logps/chosen": -1.0340476036071777, | |
| "logps/rejected": -1.1598145961761475, | |
| "loss": 1.1976, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0680952072143555, | |
| "rewards/margins": 0.251534104347229, | |
| "rewards/rejected": -2.319629192352295, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 1.2536023054755043, | |
| "grad_norm": 17.23248745457562, | |
| "learning_rate": 3.61335243028146e-08, | |
| "logits/chosen": -2.0114941596984863, | |
| "logits/rejected": -2.016153573989868, | |
| "logps/chosen": -1.092045783996582, | |
| "logps/rejected": -1.2228668928146362, | |
| "loss": 1.1916, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.184091567993164, | |
| "rewards/margins": 0.2616419494152069, | |
| "rewards/rejected": -2.4457337856292725, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 1.260806916426513, | |
| "grad_norm": 18.437527072676268, | |
| "learning_rate": 3.5945461061099736e-08, | |
| "logits/chosen": -1.972100853919983, | |
| "logits/rejected": -1.9586093425750732, | |
| "logps/chosen": -1.0443975925445557, | |
| "logps/rejected": -1.1218501329421997, | |
| "loss": 1.2706, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.0887951850891113, | |
| "rewards/margins": 0.15490522980690002, | |
| "rewards/rejected": -2.2437002658843994, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.2680115273775217, | |
| "grad_norm": 19.695402848445642, | |
| "learning_rate": 3.5756628398668446e-08, | |
| "logits/chosen": -2.0573649406433105, | |
| "logits/rejected": -2.0625429153442383, | |
| "logps/chosen": -1.1325814723968506, | |
| "logps/rejected": -1.232399821281433, | |
| "loss": 1.2403, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.265162944793701, | |
| "rewards/margins": 0.19963672757148743, | |
| "rewards/rejected": -2.464799642562866, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 1.2752161383285303, | |
| "grad_norm": 17.758331420145563, | |
| "learning_rate": 3.556703958967716e-08, | |
| "logits/chosen": -2.041581630706787, | |
| "logits/rejected": -2.036958694458008, | |
| "logps/chosen": -1.0513150691986084, | |
| "logps/rejected": -1.1853423118591309, | |
| "loss": 1.1882, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.102630138397217, | |
| "rewards/margins": 0.2680542469024658, | |
| "rewards/rejected": -2.3706846237182617, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 1.282420749279539, | |
| "grad_norm": 24.11832525210908, | |
| "learning_rate": 3.5376707961436297e-08, | |
| "logits/chosen": -2.025054454803467, | |
| "logits/rejected": -2.019120693206787, | |
| "logps/chosen": -1.1408239603042603, | |
| "logps/rejected": -1.202470064163208, | |
| "loss": 1.2726, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.2816479206085205, | |
| "rewards/margins": 0.12329187244176865, | |
| "rewards/rejected": -2.404940128326416, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 1.2896253602305476, | |
| "grad_norm": 12.89708045158757, | |
| "learning_rate": 3.51856468934734e-08, | |
| "logits/chosen": -1.9773098230361938, | |
| "logits/rejected": -1.9786754846572876, | |
| "logps/chosen": -0.9762522578239441, | |
| "logps/rejected": -1.0697864294052124, | |
| "loss": 1.2234, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.9525045156478882, | |
| "rewards/margins": 0.18706828355789185, | |
| "rewards/rejected": -2.139572858810425, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 1.2968299711815563, | |
| "grad_norm": 20.10190857160128, | |
| "learning_rate": 3.499386981659262e-08, | |
| "logits/chosen": -2.0595974922180176, | |
| "logits/rejected": -2.0540311336517334, | |
| "logps/chosen": -1.0190843343734741, | |
| "logps/rejected": -1.208898901939392, | |
| "loss": 1.1243, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.0381686687469482, | |
| "rewards/margins": 0.3796289563179016, | |
| "rewards/rejected": -2.417797803878784, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.304034582132565, | |
| "grad_norm": 20.878615577501385, | |
| "learning_rate": 3.480139021193057e-08, | |
| "logits/chosen": -1.9839977025985718, | |
| "logits/rejected": -1.9858938455581665, | |
| "logps/chosen": -0.9964865446090698, | |
| "logps/rejected": -1.1170662641525269, | |
| "loss": 1.212, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.9929730892181396, | |
| "rewards/margins": 0.24115952849388123, | |
| "rewards/rejected": -2.2341325283050537, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 1.3112391930835736, | |
| "grad_norm": 28.47013732688272, | |
| "learning_rate": 3.4608221610008666e-08, | |
| "logits/chosen": -2.0153214931488037, | |
| "logits/rejected": -2.010758876800537, | |
| "logps/chosen": -0.9736091494560242, | |
| "logps/rejected": -1.120499849319458, | |
| "loss": 1.1707, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -1.9472182989120483, | |
| "rewards/margins": 0.29378125071525574, | |
| "rewards/rejected": -2.240999698638916, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 1.318443804034582, | |
| "grad_norm": 15.221657015785182, | |
| "learning_rate": 3.4414377589782e-08, | |
| "logits/chosen": -1.9868896007537842, | |
| "logits/rejected": -1.9957456588745117, | |
| "logps/chosen": -1.0180634260177612, | |
| "logps/rejected": -1.150467872619629, | |
| "loss": 1.1966, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.0361268520355225, | |
| "rewards/margins": 0.26480910181999207, | |
| "rewards/rejected": -2.300935745239258, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 1.3256484149855907, | |
| "grad_norm": 18.190653029469026, | |
| "learning_rate": 3.4219871777684745e-08, | |
| "logits/chosen": -1.9982116222381592, | |
| "logits/rejected": -1.9859825372695923, | |
| "logps/chosen": -0.9929243922233582, | |
| "logps/rejected": -1.1142441034317017, | |
| "loss": 1.2076, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.9858487844467163, | |
| "rewards/margins": 0.24263925850391388, | |
| "rewards/rejected": -2.2284882068634033, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 1.3328530259365994, | |
| "grad_norm": 17.791029774645512, | |
| "learning_rate": 3.4024717846672364e-08, | |
| "logits/chosen": -2.0318691730499268, | |
| "logits/rejected": -2.025087833404541, | |
| "logps/chosen": -0.9934013485908508, | |
| "logps/rejected": -1.1215975284576416, | |
| "loss": 1.1959, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -1.9868026971817017, | |
| "rewards/margins": 0.25639256834983826, | |
| "rewards/rejected": -2.243195056915283, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.340057636887608, | |
| "grad_norm": 17.7783196169273, | |
| "learning_rate": 3.382892951526036e-08, | |
| "logits/chosen": -2.0219979286193848, | |
| "logits/rejected": -2.0191798210144043, | |
| "logps/chosen": -1.0518848896026611, | |
| "logps/rejected": -1.20078444480896, | |
| "loss": 1.1628, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.1037697792053223, | |
| "rewards/margins": 0.2977990508079529, | |
| "rewards/rejected": -2.40156888961792, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 1.3472622478386167, | |
| "grad_norm": 20.2114199388819, | |
| "learning_rate": 3.3632520546559974e-08, | |
| "logits/chosen": -1.9855273962020874, | |
| "logits/rejected": -1.9737205505371094, | |
| "logps/chosen": -0.926679253578186, | |
| "logps/rejected": -1.0955464839935303, | |
| "loss": 1.1271, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.853358507156372, | |
| "rewards/margins": 0.3377344310283661, | |
| "rewards/rejected": -2.1910929679870605, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 1.3544668587896254, | |
| "grad_norm": 19.579421951203443, | |
| "learning_rate": 3.34355047473107e-08, | |
| "logits/chosen": -1.9991194009780884, | |
| "logits/rejected": -1.9949671030044556, | |
| "logps/chosen": -1.0290135145187378, | |
| "logps/rejected": -1.1186621189117432, | |
| "loss": 1.2445, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.0580270290374756, | |
| "rewards/margins": 0.1792970895767212, | |
| "rewards/rejected": -2.2373242378234863, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 1.361671469740634, | |
| "grad_norm": 22.936331468503273, | |
| "learning_rate": 3.323789596690971e-08, | |
| "logits/chosen": -1.966144323348999, | |
| "logits/rejected": -1.9670454263687134, | |
| "logps/chosen": -1.0209920406341553, | |
| "logps/rejected": -1.1551681756973267, | |
| "loss": 1.1787, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.0419840812683105, | |
| "rewards/margins": 0.26835212111473083, | |
| "rewards/rejected": -2.3103363513946533, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 1.3688760806916427, | |
| "grad_norm": 15.801647380635032, | |
| "learning_rate": 3.303970809643828e-08, | |
| "logits/chosen": -1.998286247253418, | |
| "logits/rejected": -2.0028045177459717, | |
| "logps/chosen": -1.0353937149047852, | |
| "logps/rejected": -1.1643174886703491, | |
| "loss": 1.1927, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0707874298095703, | |
| "rewards/margins": 0.25784778594970703, | |
| "rewards/rejected": -2.3286349773406982, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.3760806916426513, | |
| "grad_norm": 20.693514419325513, | |
| "learning_rate": 3.2840955067685356e-08, | |
| "logits/chosen": -2.031480312347412, | |
| "logits/rejected": -2.035548686981201, | |
| "logps/chosen": -1.0550123453140259, | |
| "logps/rejected": -1.2029016017913818, | |
| "loss": 1.1631, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.1100246906280518, | |
| "rewards/margins": 0.29577863216400146, | |
| "rewards/rejected": -2.4058032035827637, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 1.38328530259366, | |
| "grad_norm": 16.858093329362955, | |
| "learning_rate": 3.264165085216817e-08, | |
| "logits/chosen": -2.038879871368408, | |
| "logits/rejected": -2.0388429164886475, | |
| "logps/chosen": -0.9352089166641235, | |
| "logps/rejected": -1.10355544090271, | |
| "loss": 1.1401, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.870417833328247, | |
| "rewards/margins": 0.33669325709342957, | |
| "rewards/rejected": -2.20711088180542, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 1.3904899135446687, | |
| "grad_norm": 18.690349536010206, | |
| "learning_rate": 3.244180946015008e-08, | |
| "logits/chosen": -1.966835618019104, | |
| "logits/rejected": -1.967462182044983, | |
| "logps/chosen": -1.0351486206054688, | |
| "logps/rejected": -1.0991723537445068, | |
| "loss": 1.273, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.0702972412109375, | |
| "rewards/margins": 0.12804751098155975, | |
| "rewards/rejected": -2.1983447074890137, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 1.397694524495677, | |
| "grad_norm": 15.348372078288971, | |
| "learning_rate": 3.224144493965578e-08, | |
| "logits/chosen": -2.0523886680603027, | |
| "logits/rejected": -2.0558857917785645, | |
| "logps/chosen": -0.9908173680305481, | |
| "logps/rejected": -1.0954809188842773, | |
| "loss": 1.2177, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -1.9816347360610962, | |
| "rewards/margins": 0.20932729542255402, | |
| "rewards/rejected": -2.1909618377685547, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 1.4048991354466858, | |
| "grad_norm": 17.879874010257755, | |
| "learning_rate": 3.204057137548371e-08, | |
| "logits/chosen": -2.014993667602539, | |
| "logits/rejected": -2.0096094608306885, | |
| "logps/chosen": -0.9776601791381836, | |
| "logps/rejected": -1.0827131271362305, | |
| "loss": 1.2165, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.9553203582763672, | |
| "rewards/margins": 0.2101059854030609, | |
| "rewards/rejected": -2.165426254272461, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.4121037463976944, | |
| "grad_norm": 19.498418734777132, | |
| "learning_rate": 3.183920288821597e-08, | |
| "logits/chosen": -1.9974403381347656, | |
| "logits/rejected": -1.9938675165176392, | |
| "logps/chosen": -1.002251386642456, | |
| "logps/rejected": -1.1633012294769287, | |
| "loss": 1.1473, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.004502773284912, | |
| "rewards/margins": 0.32209956645965576, | |
| "rewards/rejected": -2.3266024589538574, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 1.419308357348703, | |
| "grad_norm": 23.526801289262714, | |
| "learning_rate": 3.1637353633225735e-08, | |
| "logits/chosen": -2.0382392406463623, | |
| "logits/rejected": -2.0321145057678223, | |
| "logps/chosen": -1.0285365581512451, | |
| "logps/rejected": -1.1749916076660156, | |
| "loss": 1.1708, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0570731163024902, | |
| "rewards/margins": 0.2929099202156067, | |
| "rewards/rejected": -2.3499832153320312, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 1.4265129682997117, | |
| "grad_norm": 19.581143803282398, | |
| "learning_rate": 3.143503779968213e-08, | |
| "logits/chosen": -2.0113444328308105, | |
| "logits/rejected": -2.011580467224121, | |
| "logps/chosen": -1.0159164667129517, | |
| "logps/rejected": -1.1518559455871582, | |
| "loss": 1.196, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0318329334259033, | |
| "rewards/margins": 0.2718789875507355, | |
| "rewards/rejected": -2.3037118911743164, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.4337175792507204, | |
| "grad_norm": 18.05404403193421, | |
| "learning_rate": 3.1232269609552875e-08, | |
| "logits/chosen": -1.9945173263549805, | |
| "logits/rejected": -1.9919058084487915, | |
| "logps/chosen": -0.9980915188789368, | |
| "logps/rejected": -1.1186559200286865, | |
| "loss": 1.2014, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.9961830377578735, | |
| "rewards/margins": 0.24112899601459503, | |
| "rewards/rejected": -2.237311840057373, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 1.440922190201729, | |
| "grad_norm": 16.09307467422962, | |
| "learning_rate": 3.102906331660444e-08, | |
| "logits/chosen": -2.0536019802093506, | |
| "logits/rejected": -2.045327663421631, | |
| "logps/chosen": -0.9929038882255554, | |
| "logps/rejected": -1.1643650531768799, | |
| "loss": 1.1353, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.9858077764511108, | |
| "rewards/margins": 0.34292247891426086, | |
| "rewards/rejected": -2.3287301063537598, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.4481268011527377, | |
| "grad_norm": 16.081259631225404, | |
| "learning_rate": 3.082543320540015e-08, | |
| "logits/chosen": -1.9962193965911865, | |
| "logits/rejected": -1.9890375137329102, | |
| "logps/chosen": -1.0065879821777344, | |
| "logps/rejected": -1.1499404907226562, | |
| "loss": 1.1679, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0131759643554688, | |
| "rewards/margins": 0.2867050766944885, | |
| "rewards/rejected": -2.2998809814453125, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 1.4553314121037464, | |
| "grad_norm": 18.028275293304183, | |
| "learning_rate": 3.062139359029599e-08, | |
| "logits/chosen": -2.031736373901367, | |
| "logits/rejected": -2.0316202640533447, | |
| "logps/chosen": -1.0291239023208618, | |
| "logps/rejected": -1.1133326292037964, | |
| "loss": 1.2476, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -2.0582478046417236, | |
| "rewards/margins": 0.16841746866703033, | |
| "rewards/rejected": -2.2266652584075928, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 1.462536023054755, | |
| "grad_norm": 18.739986191205507, | |
| "learning_rate": 3.041695881443437e-08, | |
| "logits/chosen": -2.0472700595855713, | |
| "logits/rejected": -2.0425424575805664, | |
| "logps/chosen": -0.9730477333068848, | |
| "logps/rejected": -1.1086480617523193, | |
| "loss": 1.1771, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -1.9460954666137695, | |
| "rewards/margins": 0.271200567483902, | |
| "rewards/rejected": -2.2172961235046387, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 1.4697406340057637, | |
| "grad_norm": 22.13845084834241, | |
| "learning_rate": 3.0212143248735886e-08, | |
| "logits/chosen": -2.0294270515441895, | |
| "logits/rejected": -2.029846668243408, | |
| "logps/chosen": -0.9991294741630554, | |
| "logps/rejected": -1.1360986232757568, | |
| "loss": 1.1761, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.9982589483261108, | |
| "rewards/margins": 0.27393826842308044, | |
| "rewards/rejected": -2.2721972465515137, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 1.4769452449567724, | |
| "grad_norm": 19.822466302624346, | |
| "learning_rate": 3.0006961290889077e-08, | |
| "logits/chosen": -2.0190889835357666, | |
| "logits/rejected": -2.0099661350250244, | |
| "logps/chosen": -1.1185331344604492, | |
| "logps/rejected": -1.286892056465149, | |
| "loss": 1.1647, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.2370662689208984, | |
| "rewards/margins": 0.33671754598617554, | |
| "rewards/rejected": -2.573784112930298, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.484149855907781, | |
| "grad_norm": 21.925715491881135, | |
| "learning_rate": 2.980142736433833e-08, | |
| "logits/chosen": -2.01119327545166, | |
| "logits/rejected": -2.004316806793213, | |
| "logps/chosen": -1.0309051275253296, | |
| "logps/rejected": -1.0949158668518066, | |
| "loss": 1.2751, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -2.061810255050659, | |
| "rewards/margins": 0.12802138924598694, | |
| "rewards/rejected": -2.1898317337036133, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 1.4913544668587897, | |
| "grad_norm": 24.46772736032293, | |
| "learning_rate": 2.9595555917269997e-08, | |
| "logits/chosen": -2.03961181640625, | |
| "logits/rejected": -2.0247714519500732, | |
| "logps/chosen": -1.1411329507827759, | |
| "logps/rejected": -1.2373685836791992, | |
| "loss": 1.2153, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.2822659015655518, | |
| "rewards/margins": 0.19247153401374817, | |
| "rewards/rejected": -2.4747371673583984, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 1.4985590778097984, | |
| "grad_norm": 18.449968646671344, | |
| "learning_rate": 2.9389361421596725e-08, | |
| "logits/chosen": -1.9533805847167969, | |
| "logits/rejected": -1.9556515216827393, | |
| "logps/chosen": -1.0595835447311401, | |
| "logps/rejected": -1.1933083534240723, | |
| "loss": 1.1832, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.1191670894622803, | |
| "rewards/margins": 0.2674497365951538, | |
| "rewards/rejected": -2.3866167068481445, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.505763688760807, | |
| "grad_norm": 20.34455177562933, | |
| "learning_rate": 2.9182858371940126e-08, | |
| "logits/chosen": -2.0372543334960938, | |
| "logits/rejected": -2.031832218170166, | |
| "logps/chosen": -1.0473064184188843, | |
| "logps/rejected": -1.1757621765136719, | |
| "loss": 1.188, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0946128368377686, | |
| "rewards/margins": 0.25691163539886475, | |
| "rewards/rejected": -2.3515243530273438, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 1.5129682997118157, | |
| "grad_norm": 18.90759740416456, | |
| "learning_rate": 2.8976061284611908e-08, | |
| "logits/chosen": -1.9889026880264282, | |
| "logits/rejected": -1.9977174997329712, | |
| "logps/chosen": -0.9364235997200012, | |
| "logps/rejected": -1.0655431747436523, | |
| "loss": 1.1944, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.8728471994400024, | |
| "rewards/margins": 0.2582393288612366, | |
| "rewards/rejected": -2.1310863494873047, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.5201729106628243, | |
| "grad_norm": 21.434032214198695, | |
| "learning_rate": 2.8768984696593384e-08, | |
| "logits/chosen": -1.9844554662704468, | |
| "logits/rejected": -1.974907636642456, | |
| "logps/chosen": -1.0168864727020264, | |
| "logps/rejected": -1.1343626976013184, | |
| "loss": 1.2154, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.0337729454040527, | |
| "rewards/margins": 0.23495233058929443, | |
| "rewards/rejected": -2.2687253952026367, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 1.527377521613833, | |
| "grad_norm": 18.047284778863265, | |
| "learning_rate": 2.8561643164513637e-08, | |
| "logits/chosen": -1.9067440032958984, | |
| "logits/rejected": -1.9029529094696045, | |
| "logps/chosen": -1.0492842197418213, | |
| "logps/rejected": -1.1676268577575684, | |
| "loss": 1.2013, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.0985684394836426, | |
| "rewards/margins": 0.23668520152568817, | |
| "rewards/rejected": -2.3352537155151367, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 1.5345821325648417, | |
| "grad_norm": 18.95635925899202, | |
| "learning_rate": 2.8354051263626227e-08, | |
| "logits/chosen": -1.9887434244155884, | |
| "logits/rejected": -1.994476556777954, | |
| "logps/chosen": -1.0596574544906616, | |
| "logps/rejected": -1.1733436584472656, | |
| "loss": 1.206, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.1193149089813232, | |
| "rewards/margins": 0.22737233340740204, | |
| "rewards/rejected": -2.3466873168945312, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 1.54178674351585, | |
| "grad_norm": 19.584229312796637, | |
| "learning_rate": 2.8146223586784573e-08, | |
| "logits/chosen": -1.9805008172988892, | |
| "logits/rejected": -1.9726234674453735, | |
| "logps/chosen": -1.0646823644638062, | |
| "logps/rejected": -1.1987252235412598, | |
| "loss": 1.1873, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.1293647289276123, | |
| "rewards/margins": 0.26808565855026245, | |
| "rewards/rejected": -2.3974504470825195, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 1.5489913544668588, | |
| "grad_norm": 25.50415369546022, | |
| "learning_rate": 2.7938174743416205e-08, | |
| "logits/chosen": -1.9369735717773438, | |
| "logits/rejected": -1.933683156967163, | |
| "logps/chosen": -1.050445795059204, | |
| "logps/rejected": -1.1607348918914795, | |
| "loss": 1.2135, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.100891590118408, | |
| "rewards/margins": 0.220577672123909, | |
| "rewards/rejected": -2.321469783782959, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.5561959654178674, | |
| "grad_norm": 19.684619038178205, | |
| "learning_rate": 2.7729919358495728e-08, | |
| "logits/chosen": -2.005277395248413, | |
| "logits/rejected": -2.0062077045440674, | |
| "logps/chosen": -1.1121950149536133, | |
| "logps/rejected": -1.19098699092865, | |
| "loss": 1.2586, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.2243900299072266, | |
| "rewards/margins": 0.1575840413570404, | |
| "rewards/rejected": -2.3819739818573, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 1.563400576368876, | |
| "grad_norm": 19.293166467927325, | |
| "learning_rate": 2.7521472071516772e-08, | |
| "logits/chosen": -2.0027170181274414, | |
| "logits/rejected": -2.0016961097717285, | |
| "logps/chosen": -0.9449695348739624, | |
| "logps/rejected": -1.0605154037475586, | |
| "loss": 1.2076, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.8899390697479248, | |
| "rewards/margins": 0.2310914546251297, | |
| "rewards/rejected": -2.121030807495117, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 1.5706051873198847, | |
| "grad_norm": 22.062496687144794, | |
| "learning_rate": 2.731284753546289e-08, | |
| "logits/chosen": -1.9856891632080078, | |
| "logits/rejected": -1.9836666584014893, | |
| "logps/chosen": -1.081839919090271, | |
| "logps/rejected": -1.2224990129470825, | |
| "loss": 1.1741, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.163679838180542, | |
| "rewards/margins": 0.2813180387020111, | |
| "rewards/rejected": -2.444998025894165, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 1.5778097982708934, | |
| "grad_norm": 21.803351526445823, | |
| "learning_rate": 2.710406041577751e-08, | |
| "logits/chosen": -2.04976224899292, | |
| "logits/rejected": -2.0463500022888184, | |
| "logps/chosen": -1.0325794219970703, | |
| "logps/rejected": -1.1858645677566528, | |
| "loss": 1.1631, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0651588439941406, | |
| "rewards/margins": 0.3065701127052307, | |
| "rewards/rejected": -2.3717291355133057, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 1.585014409221902, | |
| "grad_norm": 18.0281741107113, | |
| "learning_rate": 2.6895125389333017e-08, | |
| "logits/chosen": -2.0131421089172363, | |
| "logits/rejected": -2.0089025497436523, | |
| "logps/chosen": -1.0270203351974487, | |
| "logps/rejected": -1.177971363067627, | |
| "loss": 1.1622, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.0540406703948975, | |
| "rewards/margins": 0.3019018769264221, | |
| "rewards/rejected": -2.355942726135254, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.5922190201729105, | |
| "grad_norm": 17.129921710950377, | |
| "learning_rate": 2.6686057143399028e-08, | |
| "logits/chosen": -2.010429620742798, | |
| "logits/rejected": -2.0119571685791016, | |
| "logps/chosen": -1.0614731311798096, | |
| "logps/rejected": -1.1593468189239502, | |
| "loss": 1.2433, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.122946262359619, | |
| "rewards/margins": 0.1957472264766693, | |
| "rewards/rejected": -2.3186936378479004, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 1.5994236311239192, | |
| "grad_norm": 19.402506516811066, | |
| "learning_rate": 2.647687037460996e-08, | |
| "logits/chosen": -2.0160350799560547, | |
| "logits/rejected": -2.0153958797454834, | |
| "logps/chosen": -1.0873758792877197, | |
| "logps/rejected": -1.2827941179275513, | |
| "loss": 1.1246, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.1747517585754395, | |
| "rewards/margins": 0.3908364176750183, | |
| "rewards/rejected": -2.5655882358551025, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 1.6066282420749278, | |
| "grad_norm": 20.370963061014333, | |
| "learning_rate": 2.626757978793187e-08, | |
| "logits/chosen": -2.0244648456573486, | |
| "logits/rejected": -2.0181853771209717, | |
| "logps/chosen": -1.0852004289627075, | |
| "logps/rejected": -1.2089064121246338, | |
| "loss": 1.2036, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.170400857925415, | |
| "rewards/margins": 0.24741193652153015, | |
| "rewards/rejected": -2.4178128242492676, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 1.6138328530259365, | |
| "grad_norm": 23.538795309630903, | |
| "learning_rate": 2.6058200095628797e-08, | |
| "logits/chosen": -1.9968335628509521, | |
| "logits/rejected": -2.000123977661133, | |
| "logps/chosen": -0.9178045392036438, | |
| "logps/rejected": -1.086455225944519, | |
| "loss": 1.1446, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -1.8356090784072876, | |
| "rewards/margins": 0.3373013734817505, | |
| "rewards/rejected": -2.172910451889038, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 1.6210374639769451, | |
| "grad_norm": 18.7834477811749, | |
| "learning_rate": 2.584874601622854e-08, | |
| "logits/chosen": -2.0577359199523926, | |
| "logits/rejected": -2.0486464500427246, | |
| "logps/chosen": -1.0842779874801636, | |
| "logps/rejected": -1.2169630527496338, | |
| "loss": 1.2055, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.168555974960327, | |
| "rewards/margins": 0.26537027955055237, | |
| "rewards/rejected": -2.4339261054992676, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.6282420749279538, | |
| "grad_norm": 21.50129735883824, | |
| "learning_rate": 2.5639232273487993e-08, | |
| "logits/chosen": -1.9792057275772095, | |
| "logits/rejected": -1.9694305658340454, | |
| "logps/chosen": -0.9786102175712585, | |
| "logps/rejected": -1.0999042987823486, | |
| "loss": 1.2022, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.957220435142517, | |
| "rewards/margins": 0.24258823692798615, | |
| "rewards/rejected": -2.1998085975646973, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 1.6354466858789625, | |
| "grad_norm": 20.836699972853967, | |
| "learning_rate": 2.5429673595358142e-08, | |
| "logits/chosen": -2.0185582637786865, | |
| "logits/rejected": -2.0170459747314453, | |
| "logps/chosen": -1.043128490447998, | |
| "logps/rejected": -1.165533185005188, | |
| "loss": 1.1967, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.086256980895996, | |
| "rewards/margins": 0.24480919539928436, | |
| "rewards/rejected": -2.331066370010376, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 1.6426512968299711, | |
| "grad_norm": 23.73080611195804, | |
| "learning_rate": 2.5220084712948764e-08, | |
| "logits/chosen": -1.9826107025146484, | |
| "logits/rejected": -1.9717817306518555, | |
| "logps/chosen": -1.1198116540908813, | |
| "logps/rejected": -1.2381196022033691, | |
| "loss": 1.1906, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.2396233081817627, | |
| "rewards/margins": 0.23661574721336365, | |
| "rewards/rejected": -2.4762392044067383, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 1.6498559077809798, | |
| "grad_norm": 19.691578624312058, | |
| "learning_rate": 2.5010480359492838e-08, | |
| "logits/chosen": -1.9650003910064697, | |
| "logits/rejected": -1.9621715545654297, | |
| "logps/chosen": -1.0505197048187256, | |
| "logps/rejected": -1.1106680631637573, | |
| "loss": 1.2861, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.101039409637451, | |
| "rewards/margins": 0.12029679119586945, | |
| "rewards/rejected": -2.2213361263275146, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 1.6570605187319885, | |
| "grad_norm": 21.239713120458195, | |
| "learning_rate": 2.480087526931091e-08, | |
| "logits/chosen": -2.0088658332824707, | |
| "logits/rejected": -1.9966083765029907, | |
| "logps/chosen": -1.0031955242156982, | |
| "logps/rejected": -1.1170064210891724, | |
| "loss": 1.2166, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.0063910484313965, | |
| "rewards/margins": 0.22762183845043182, | |
| "rewards/rejected": -2.2340128421783447, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.6642651296829971, | |
| "grad_norm": 19.4121166793283, | |
| "learning_rate": 2.4591284176775326e-08, | |
| "logits/chosen": -1.9742721319198608, | |
| "logits/rejected": -1.9704244136810303, | |
| "logps/chosen": -1.07572340965271, | |
| "logps/rejected": -1.159128189086914, | |
| "loss": 1.2565, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.15144681930542, | |
| "rewards/margins": 0.16680975258350372, | |
| "rewards/rejected": -2.318256378173828, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 1.6714697406340058, | |
| "grad_norm": 21.976079747343572, | |
| "learning_rate": 2.4381721815274443e-08, | |
| "logits/chosen": -2.0400168895721436, | |
| "logits/rejected": -2.0402297973632812, | |
| "logps/chosen": -1.019706130027771, | |
| "logps/rejected": -1.1512401103973389, | |
| "loss": 1.1928, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.039412260055542, | |
| "rewards/margins": 0.26306766271591187, | |
| "rewards/rejected": -2.3024802207946777, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 1.6786743515850144, | |
| "grad_norm": 19.583839102475277, | |
| "learning_rate": 2.4172202916176936e-08, | |
| "logits/chosen": -2.0487046241760254, | |
| "logits/rejected": -2.0510191917419434, | |
| "logps/chosen": -0.9676868319511414, | |
| "logps/rejected": -1.1354566812515259, | |
| "loss": 1.1567, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -1.9353736639022827, | |
| "rewards/margins": 0.3355395197868347, | |
| "rewards/rejected": -2.2709133625030518, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 1.685878962536023, | |
| "grad_norm": 19.19182662272249, | |
| "learning_rate": 2.3962742207796268e-08, | |
| "logits/chosen": -1.9858787059783936, | |
| "logits/rejected": -1.9837026596069336, | |
| "logps/chosen": -0.9570139050483704, | |
| "logps/rejected": -1.1192009449005127, | |
| "loss": 1.1603, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.9140278100967407, | |
| "rewards/margins": 0.32437413930892944, | |
| "rewards/rejected": -2.2384018898010254, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 1.6930835734870318, | |
| "grad_norm": 22.691977894194924, | |
| "learning_rate": 2.3753354414355334e-08, | |
| "logits/chosen": -1.9461901187896729, | |
| "logits/rejected": -1.9355932474136353, | |
| "logps/chosen": -1.0649149417877197, | |
| "logps/rejected": -1.1817948818206787, | |
| "loss": 1.213, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.1298298835754395, | |
| "rewards/margins": 0.2337600290775299, | |
| "rewards/rejected": -2.3635897636413574, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.7002881844380404, | |
| "grad_norm": 18.59945891396093, | |
| "learning_rate": 2.3544054254951408e-08, | |
| "logits/chosen": -1.9878515005111694, | |
| "logits/rejected": -1.9792087078094482, | |
| "logps/chosen": -0.9375497102737427, | |
| "logps/rejected": -1.1345303058624268, | |
| "loss": 1.114, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -1.8750994205474854, | |
| "rewards/margins": 0.393961638212204, | |
| "rewards/rejected": -2.2690606117248535, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 1.707492795389049, | |
| "grad_norm": 18.203541895462912, | |
| "learning_rate": 2.3334856442521435e-08, | |
| "logits/chosen": -2.0370235443115234, | |
| "logits/rejected": -2.0295424461364746, | |
| "logps/chosen": -1.0964655876159668, | |
| "logps/rejected": -1.166515588760376, | |
| "loss": 1.27, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -2.1929311752319336, | |
| "rewards/margins": 0.1401001363992691, | |
| "rewards/rejected": -2.333031177520752, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 1.7146974063400577, | |
| "grad_norm": 19.123513495613718, | |
| "learning_rate": 2.3125775682807826e-08, | |
| "logits/chosen": -2.0507147312164307, | |
| "logits/rejected": -2.0506680011749268, | |
| "logps/chosen": -1.1658060550689697, | |
| "logps/rejected": -1.2665237188339233, | |
| "loss": 1.2309, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.3316121101379395, | |
| "rewards/margins": 0.2014356553554535, | |
| "rewards/rejected": -2.5330474376678467, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 1.7219020172910664, | |
| "grad_norm": 20.583955091856193, | |
| "learning_rate": 2.291682667332464e-08, | |
| "logits/chosen": -2.0658912658691406, | |
| "logits/rejected": -2.0607848167419434, | |
| "logps/chosen": -1.0484416484832764, | |
| "logps/rejected": -1.1794416904449463, | |
| "loss": 1.1918, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.0968832969665527, | |
| "rewards/margins": 0.2620001435279846, | |
| "rewards/rejected": -2.3588833808898926, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 1.729106628242075, | |
| "grad_norm": 15.255925002553854, | |
| "learning_rate": 2.2708024102324454e-08, | |
| "logits/chosen": -2.0251784324645996, | |
| "logits/rejected": -2.0195064544677734, | |
| "logps/chosen": -1.0335527658462524, | |
| "logps/rejected": -1.2097657918930054, | |
| "loss": 1.1498, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.067105531692505, | |
| "rewards/margins": 0.35242635011672974, | |
| "rewards/rejected": -2.4195315837860107, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.7363112391930837, | |
| "grad_norm": 22.44593573299748, | |
| "learning_rate": 2.2499382647765797e-08, | |
| "logits/chosen": -2.0198001861572266, | |
| "logits/rejected": -2.016092538833618, | |
| "logps/chosen": -1.0722882747650146, | |
| "logps/rejected": -1.161583662033081, | |
| "loss": 1.2463, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.1445765495300293, | |
| "rewards/margins": 0.17859075963497162, | |
| "rewards/rejected": -2.323167324066162, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 1.7435158501440924, | |
| "grad_norm": 20.996477598226324, | |
| "learning_rate": 2.2290916976281427e-08, | |
| "logits/chosen": -1.997984528541565, | |
| "logits/rejected": -1.991624116897583, | |
| "logps/chosen": -0.9992947578430176, | |
| "logps/rejected": -1.1312335729599, | |
| "loss": 1.2149, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -1.9985895156860352, | |
| "rewards/margins": 0.26387742161750793, | |
| "rewards/rejected": -2.2624671459198, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 1.7507204610951008, | |
| "grad_norm": 18.145146158512926, | |
| "learning_rate": 2.2082641742147238e-08, | |
| "logits/chosen": -1.9863611459732056, | |
| "logits/rejected": -1.9797251224517822, | |
| "logps/chosen": -1.0165367126464844, | |
| "logps/rejected": -1.2077693939208984, | |
| "loss": 1.115, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -2.0330734252929688, | |
| "rewards/margins": 0.38246554136276245, | |
| "rewards/rejected": -2.415538787841797, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 1.7579250720461095, | |
| "grad_norm": 20.511354788346416, | |
| "learning_rate": 2.1874571586252177e-08, | |
| "logits/chosen": -2.0291788578033447, | |
| "logits/rejected": -2.0222790241241455, | |
| "logps/chosen": -1.0278832912445068, | |
| "logps/rejected": -1.1068981885910034, | |
| "loss": 1.256, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0557665824890137, | |
| "rewards/margins": 0.15802964568138123, | |
| "rewards/rejected": -2.213796377182007, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 1.7651296829971181, | |
| "grad_norm": 20.78736849578736, | |
| "learning_rate": 2.1666721135069037e-08, | |
| "logits/chosen": -2.022594928741455, | |
| "logits/rejected": -2.019284725189209, | |
| "logps/chosen": -1.1104170083999634, | |
| "logps/rejected": -1.2043354511260986, | |
| "loss": 1.2436, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.2208340167999268, | |
| "rewards/margins": 0.18783698976039886, | |
| "rewards/rejected": -2.4086709022521973, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.7723342939481268, | |
| "grad_norm": 15.559026450288725, | |
| "learning_rate": 2.145910499962628e-08, | |
| "logits/chosen": -2.0644400119781494, | |
| "logits/rejected": -2.0565133094787598, | |
| "logps/chosen": -0.9585247039794922, | |
| "logps/rejected": -1.101301908493042, | |
| "loss": 1.1824, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9170494079589844, | |
| "rewards/margins": 0.28555426001548767, | |
| "rewards/rejected": -2.202603816986084, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 1.7795389048991355, | |
| "grad_norm": 23.88258329458798, | |
| "learning_rate": 2.1251737774480915e-08, | |
| "logits/chosen": -2.0456204414367676, | |
| "logits/rejected": -2.036010265350342, | |
| "logps/chosen": -1.1689999103546143, | |
| "logps/rejected": -1.2592300176620483, | |
| "loss": 1.2704, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.3379998207092285, | |
| "rewards/margins": 0.18046024441719055, | |
| "rewards/rejected": -2.5184600353240967, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 1.7867435158501441, | |
| "grad_norm": 17.478397647824718, | |
| "learning_rate": 2.104463403669264e-08, | |
| "logits/chosen": -1.9978790283203125, | |
| "logits/rejected": -1.9951884746551514, | |
| "logps/chosen": -1.0451444387435913, | |
| "logps/rejected": -1.189968466758728, | |
| "loss": 1.1806, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.0902888774871826, | |
| "rewards/margins": 0.2896478772163391, | |
| "rewards/rejected": -2.379936933517456, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 1.7939481268011528, | |
| "grad_norm": 17.10039588248249, | |
| "learning_rate": 2.0837808344799028e-08, | |
| "logits/chosen": -1.9799407720565796, | |
| "logits/rejected": -1.9754537343978882, | |
| "logps/chosen": -0.9404538869857788, | |
| "logps/rejected": -1.0731335878372192, | |
| "loss": 1.1826, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.8809077739715576, | |
| "rewards/margins": 0.26535919308662415, | |
| "rewards/rejected": -2.1462671756744385, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 1.8011527377521612, | |
| "grad_norm": 18.18583469521082, | |
| "learning_rate": 2.063127523779219e-08, | |
| "logits/chosen": -1.9833685159683228, | |
| "logits/rejected": -1.9792015552520752, | |
| "logps/chosen": -1.0076165199279785, | |
| "logps/rejected": -1.1942096948623657, | |
| "loss": 1.1139, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": -2.015233039855957, | |
| "rewards/margins": 0.373186320066452, | |
| "rewards/rejected": -2.3884193897247314, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.8083573487031699, | |
| "grad_norm": 19.97417842705391, | |
| "learning_rate": 2.0425049234096737e-08, | |
| "logits/chosen": -1.9911282062530518, | |
| "logits/rejected": -1.9853017330169678, | |
| "logps/chosen": -1.0088120698928833, | |
| "logps/rejected": -1.1258007287979126, | |
| "loss": 1.2158, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.0176241397857666, | |
| "rewards/margins": 0.2339775562286377, | |
| "rewards/rejected": -2.251601457595825, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 1.8155619596541785, | |
| "grad_norm": 19.435490123277745, | |
| "learning_rate": 2.0219144830549163e-08, | |
| "logits/chosen": -1.9644883871078491, | |
| "logits/rejected": -1.9634536504745483, | |
| "logps/chosen": -1.0153688192367554, | |
| "logps/rejected": -1.161481261253357, | |
| "loss": 1.1831, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0307376384735107, | |
| "rewards/margins": 0.2922249436378479, | |
| "rewards/rejected": -2.322962522506714, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 1.8227665706051872, | |
| "grad_norm": 19.09312194813426, | |
| "learning_rate": 2.0013576501378823e-08, | |
| "logits/chosen": -1.9830167293548584, | |
| "logits/rejected": -1.9765300750732422, | |
| "logps/chosen": -1.0100529193878174, | |
| "logps/rejected": -1.144884705543518, | |
| "loss": 1.1939, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.0201058387756348, | |
| "rewards/margins": 0.2696635127067566, | |
| "rewards/rejected": -2.289769411087036, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 1.8299711815561959, | |
| "grad_norm": 20.224925594213033, | |
| "learning_rate": 1.9808358697190426e-08, | |
| "logits/chosen": -1.9704053401947021, | |
| "logits/rejected": -1.966780662536621, | |
| "logps/chosen": -0.9303935766220093, | |
| "logps/rejected": -1.0650821924209595, | |
| "loss": 1.1986, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.8607871532440186, | |
| "rewards/margins": 0.26937711238861084, | |
| "rewards/rejected": -2.130164384841919, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 1.8371757925072045, | |
| "grad_norm": 21.09688980967129, | |
| "learning_rate": 1.9603505843948214e-08, | |
| "logits/chosen": -2.01230525970459, | |
| "logits/rejected": -2.002260208129883, | |
| "logps/chosen": -0.948139488697052, | |
| "logps/rejected": -1.1198240518569946, | |
| "loss": 1.1395, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -1.896278977394104, | |
| "rewards/margins": 0.3433689475059509, | |
| "rewards/rejected": -2.2396481037139893, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.8443804034582132, | |
| "grad_norm": 20.232449119924333, | |
| "learning_rate": 1.9399032341961886e-08, | |
| "logits/chosen": -1.9766803979873657, | |
| "logits/rejected": -1.960636854171753, | |
| "logps/chosen": -0.9899090528488159, | |
| "logps/rejected": -1.0627111196517944, | |
| "loss": 1.2699, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -1.9798181056976318, | |
| "rewards/margins": 0.1456039845943451, | |
| "rewards/rejected": -2.125422239303589, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 1.8515850144092219, | |
| "grad_norm": 26.201248917968616, | |
| "learning_rate": 1.9194952564874323e-08, | |
| "logits/chosen": -2.0239641666412354, | |
| "logits/rejected": -2.0179200172424316, | |
| "logps/chosen": -1.0649644136428833, | |
| "logps/rejected": -1.2079579830169678, | |
| "loss": 1.1683, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.1299288272857666, | |
| "rewards/margins": 0.2859875559806824, | |
| "rewards/rejected": -2.4159159660339355, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 1.8587896253602305, | |
| "grad_norm": 20.644198497609576, | |
| "learning_rate": 1.8991280858651157e-08, | |
| "logits/chosen": -1.9820836782455444, | |
| "logits/rejected": -1.9764404296875, | |
| "logps/chosen": -1.0636051893234253, | |
| "logps/rejected": -1.14960777759552, | |
| "loss": 1.2503, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.1272103786468506, | |
| "rewards/margins": 0.1720050871372223, | |
| "rewards/rejected": -2.29921555519104, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 1.8659942363112392, | |
| "grad_norm": 16.854142688708556, | |
| "learning_rate": 1.8788031540572327e-08, | |
| "logits/chosen": -1.9858763217926025, | |
| "logits/rejected": -1.977818489074707, | |
| "logps/chosen": -0.9995776414871216, | |
| "logps/rejected": -1.1453144550323486, | |
| "loss": 1.1718, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -1.9991552829742432, | |
| "rewards/margins": 0.29147323966026306, | |
| "rewards/rejected": -2.2906289100646973, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 1.8731988472622478, | |
| "grad_norm": 16.996398857656907, | |
| "learning_rate": 1.858521889822565e-08, | |
| "logits/chosen": -2.0046029090881348, | |
| "logits/rejected": -2.007223129272461, | |
| "logps/chosen": -0.9735875129699707, | |
| "logps/rejected": -1.0832773447036743, | |
| "loss": 1.2233, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -1.9471750259399414, | |
| "rewards/margins": 0.21937978267669678, | |
| "rewards/rejected": -2.1665546894073486, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.8804034582132565, | |
| "grad_norm": 16.42143731996496, | |
| "learning_rate": 1.8382857188502422e-08, | |
| "logits/chosen": -1.9885772466659546, | |
| "logits/rejected": -1.9836734533309937, | |
| "logps/chosen": -0.9854310750961304, | |
| "logps/rejected": -1.1128942966461182, | |
| "loss": 1.1824, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.9708621501922607, | |
| "rewards/margins": 0.25492629408836365, | |
| "rewards/rejected": -2.2257885932922363, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 1.8876080691642652, | |
| "grad_norm": 22.507165910966208, | |
| "learning_rate": 1.8180960636595234e-08, | |
| "logits/chosen": -1.9683783054351807, | |
| "logits/rejected": -1.966205358505249, | |
| "logps/chosen": -1.0359306335449219, | |
| "logps/rejected": -1.1797659397125244, | |
| "loss": 1.1791, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.0718612670898438, | |
| "rewards/margins": 0.28767016530036926, | |
| "rewards/rejected": -2.359531879425049, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 1.8948126801152738, | |
| "grad_norm": 20.53168247865903, | |
| "learning_rate": 1.7979543434998015e-08, | |
| "logits/chosen": -2.038526773452759, | |
| "logits/rejected": -2.0433452129364014, | |
| "logps/chosen": -1.1234701871871948, | |
| "logps/rejected": -1.2116920948028564, | |
| "loss": 1.2381, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.2469403743743896, | |
| "rewards/margins": 0.17644372582435608, | |
| "rewards/rejected": -2.423384189605713, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 1.9020172910662825, | |
| "grad_norm": 26.15291556775582, | |
| "learning_rate": 1.7778619742508345e-08, | |
| "logits/chosen": -1.9968883991241455, | |
| "logits/rejected": -1.9899314641952515, | |
| "logps/chosen": -1.0930571556091309, | |
| "logps/rejected": -1.1869739294052124, | |
| "loss": 1.2543, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1861143112182617, | |
| "rewards/margins": 0.18783339858055115, | |
| "rewards/rejected": -2.373947858810425, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 1.9092219020172911, | |
| "grad_norm": 23.18245485008842, | |
| "learning_rate": 1.757820368323213e-08, | |
| "logits/chosen": -1.9929345846176147, | |
| "logits/rejected": -1.9831962585449219, | |
| "logps/chosen": -1.1062356233596802, | |
| "logps/rejected": -1.2650859355926514, | |
| "loss": 1.161, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.2124712467193604, | |
| "rewards/margins": 0.3177003860473633, | |
| "rewards/rejected": -2.5301718711853027, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.9164265129682998, | |
| "grad_norm": 22.34671676050883, | |
| "learning_rate": 1.7378309345590803e-08, | |
| "logits/chosen": -2.006321907043457, | |
| "logits/rejected": -2.015603542327881, | |
| "logps/chosen": -1.0863068103790283, | |
| "logps/rejected": -1.2286168336868286, | |
| "loss": 1.1821, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.1726136207580566, | |
| "rewards/margins": 0.2846204340457916, | |
| "rewards/rejected": -2.4572336673736572, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 1.9236311239193085, | |
| "grad_norm": 20.09934555506027, | |
| "learning_rate": 1.717895078133088e-08, | |
| "logits/chosen": -2.059466600418091, | |
| "logits/rejected": -2.0556395053863525, | |
| "logps/chosen": -1.0593311786651611, | |
| "logps/rejected": -1.2005198001861572, | |
| "loss": 1.1828, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.1186623573303223, | |
| "rewards/margins": 0.28237712383270264, | |
| "rewards/rejected": -2.4010396003723145, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 1.9308357348703171, | |
| "grad_norm": 21.005378635461394, | |
| "learning_rate": 1.698014200453624e-08, | |
| "logits/chosen": -2.0109024047851562, | |
| "logits/rejected": -2.0184760093688965, | |
| "logps/chosen": -1.031286597251892, | |
| "logps/rejected": -1.1622233390808105, | |
| "loss": 1.1776, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.062573194503784, | |
| "rewards/margins": 0.26187336444854736, | |
| "rewards/rejected": -2.324446678161621, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 1.9380403458213258, | |
| "grad_norm": 24.456240122864646, | |
| "learning_rate": 1.6781896990642964e-08, | |
| "logits/chosen": -1.9447215795516968, | |
| "logits/rejected": -1.942016839981079, | |
| "logps/chosen": -1.1477292776107788, | |
| "logps/rejected": -1.2380485534667969, | |
| "loss": 1.2441, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.2954585552215576, | |
| "rewards/margins": 0.18063834309577942, | |
| "rewards/rejected": -2.4760971069335938, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 1.9452449567723344, | |
| "grad_norm": 24.379018095612878, | |
| "learning_rate": 1.658422967545693e-08, | |
| "logits/chosen": -2.0516204833984375, | |
| "logits/rejected": -2.0385327339172363, | |
| "logps/chosen": -1.0048881769180298, | |
| "logps/rejected": -1.1195095777511597, | |
| "loss": 1.2189, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -2.0097763538360596, | |
| "rewards/margins": 0.2292429655790329, | |
| "rewards/rejected": -2.2390191555023193, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.952449567723343, | |
| "grad_norm": 20.606423235238847, | |
| "learning_rate": 1.638715395417418e-08, | |
| "logits/chosen": -2.0263454914093018, | |
| "logits/rejected": -2.024291515350342, | |
| "logps/chosen": -1.069252610206604, | |
| "logps/rejected": -1.2053518295288086, | |
| "loss": 1.1848, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.138505220413208, | |
| "rewards/margins": 0.27219831943511963, | |
| "rewards/rejected": -2.410703659057617, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 1.9596541786743515, | |
| "grad_norm": 22.535979632799137, | |
| "learning_rate": 1.619068368040416e-08, | |
| "logits/chosen": -2.024005174636841, | |
| "logits/rejected": -2.0195693969726562, | |
| "logps/chosen": -1.0005989074707031, | |
| "logps/rejected": -1.178637981414795, | |
| "loss": 1.1296, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.0011978149414062, | |
| "rewards/margins": 0.356078177690506, | |
| "rewards/rejected": -2.35727596282959, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 1.9668587896253602, | |
| "grad_norm": 17.493969053743083, | |
| "learning_rate": 1.5994832665195853e-08, | |
| "logits/chosen": -1.9611831903457642, | |
| "logits/rejected": -1.9615755081176758, | |
| "logps/chosen": -1.0340797901153564, | |
| "logps/rejected": -1.146831750869751, | |
| "loss": 1.2119, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.068159580230713, | |
| "rewards/margins": 0.22550389170646667, | |
| "rewards/rejected": -2.293663501739502, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 1.9740634005763689, | |
| "grad_norm": 20.261024993446156, | |
| "learning_rate": 1.5799614676066906e-08, | |
| "logits/chosen": -2.068851947784424, | |
| "logits/rejected": -2.065795421600342, | |
| "logps/chosen": -0.9484384655952454, | |
| "logps/rejected": -1.0868208408355713, | |
| "loss": 1.1744, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.8968769311904907, | |
| "rewards/margins": 0.2767646610736847, | |
| "rewards/rejected": -2.1736416816711426, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 1.9812680115273775, | |
| "grad_norm": 16.03971358941223, | |
| "learning_rate": 1.560504343603587e-08, | |
| "logits/chosen": -1.9830427169799805, | |
| "logits/rejected": -1.983306884765625, | |
| "logps/chosen": -1.0689435005187988, | |
| "logps/rejected": -1.224401593208313, | |
| "loss": 1.1606, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -2.1378870010375977, | |
| "rewards/margins": 0.3109160363674164, | |
| "rewards/rejected": -2.448803186416626, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.9884726224783862, | |
| "grad_norm": 18.729955235435014, | |
| "learning_rate": 1.541113262265748e-08, | |
| "logits/chosen": -2.0666756629943848, | |
| "logits/rejected": -2.0645081996917725, | |
| "logps/chosen": -1.0288206338882446, | |
| "logps/rejected": -1.1466666460037231, | |
| "loss": 1.2071, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.0576412677764893, | |
| "rewards/margins": 0.23569221794605255, | |
| "rewards/rejected": -2.2933332920074463, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 1.9956772334293948, | |
| "grad_norm": 25.946584240501473, | |
| "learning_rate": 1.5217895867061227e-08, | |
| "logits/chosen": -2.00740385055542, | |
| "logits/rejected": -2.0015203952789307, | |
| "logps/chosen": -1.0842344760894775, | |
| "logps/rejected": -1.1836035251617432, | |
| "loss": 1.2464, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.168468952178955, | |
| "rewards/margins": 0.19873787462711334, | |
| "rewards/rejected": -2.3672070503234863, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 2.0028818443804033, | |
| "grad_norm": 22.724338628633177, | |
| "learning_rate": 1.5025346752993098e-08, | |
| "logits/chosen": -1.9985382556915283, | |
| "logits/rejected": -2.000462532043457, | |
| "logps/chosen": -1.071683406829834, | |
| "logps/rejected": -1.1988445520401, | |
| "loss": 1.1999, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.143366813659668, | |
| "rewards/margins": 0.25432220101356506, | |
| "rewards/rejected": -2.3976891040802, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 2.010086455331412, | |
| "grad_norm": 23.240965924702092, | |
| "learning_rate": 1.4833498815860756e-08, | |
| "logits/chosen": -2.053358554840088, | |
| "logits/rejected": -2.055558443069458, | |
| "logps/chosen": -0.9996848106384277, | |
| "logps/rejected": -1.1848304271697998, | |
| "loss": 1.149, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9993696212768555, | |
| "rewards/margins": 0.3702912926673889, | |
| "rewards/rejected": -2.3696608543395996, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 2.0172910662824206, | |
| "grad_norm": 18.225560415881105, | |
| "learning_rate": 1.4642365541781993e-08, | |
| "logits/chosen": -1.9646400213241577, | |
| "logits/rejected": -1.9561887979507446, | |
| "logps/chosen": -1.0267730951309204, | |
| "logps/rejected": -1.1928188800811768, | |
| "loss": 1.1511, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.053546190261841, | |
| "rewards/margins": 0.3320915699005127, | |
| "rewards/rejected": -2.3856377601623535, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.0244956772334293, | |
| "grad_norm": 17.931282922261985, | |
| "learning_rate": 1.4451960366636745e-08, | |
| "logits/chosen": -2.026698589324951, | |
| "logits/rejected": -2.0378384590148926, | |
| "logps/chosen": -1.0406183004379272, | |
| "logps/rejected": -1.1752769947052002, | |
| "loss": 1.1819, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0812366008758545, | |
| "rewards/margins": 0.26931747794151306, | |
| "rewards/rejected": -2.3505539894104004, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 2.031700288184438, | |
| "grad_norm": 19.290431128690432, | |
| "learning_rate": 1.4262296675122592e-08, | |
| "logits/chosen": -2.0173158645629883, | |
| "logits/rejected": -2.0136220455169678, | |
| "logps/chosen": -1.030659556388855, | |
| "logps/rejected": -1.1910489797592163, | |
| "loss": 1.1523, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.06131911277771, | |
| "rewards/margins": 0.32077842950820923, | |
| "rewards/rejected": -2.3820979595184326, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 2.0389048991354466, | |
| "grad_norm": 17.902852888321604, | |
| "learning_rate": 1.407338779981389e-08, | |
| "logits/chosen": -1.9934546947479248, | |
| "logits/rejected": -1.9913368225097656, | |
| "logps/chosen": -0.9143481254577637, | |
| "logps/rejected": -1.0957781076431274, | |
| "loss": 1.1116, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -1.8286962509155273, | |
| "rewards/margins": 0.36285993456840515, | |
| "rewards/rejected": -2.191556215286255, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 2.0461095100864553, | |
| "grad_norm": 21.306516095869544, | |
| "learning_rate": 1.3885247020224534e-08, | |
| "logits/chosen": -2.0094637870788574, | |
| "logits/rejected": -2.004822015762329, | |
| "logps/chosen": -1.0016566514968872, | |
| "logps/rejected": -1.1311957836151123, | |
| "loss": 1.1913, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0033133029937744, | |
| "rewards/margins": 0.25907841324806213, | |
| "rewards/rejected": -2.2623915672302246, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 2.053314121037464, | |
| "grad_norm": 17.105370578566056, | |
| "learning_rate": 1.369788756187445e-08, | |
| "logits/chosen": -2.008868455886841, | |
| "logits/rejected": -2.0054876804351807, | |
| "logps/chosen": -1.0270612239837646, | |
| "logps/rejected": -1.1226084232330322, | |
| "loss": 1.2343, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.0541224479675293, | |
| "rewards/margins": 0.19109439849853516, | |
| "rewards/rejected": -2.2452168464660645, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.0605187319884726, | |
| "grad_norm": 18.695157813530198, | |
| "learning_rate": 1.3511322595359925e-08, | |
| "logits/chosen": -2.033163547515869, | |
| "logits/rejected": -2.0247857570648193, | |
| "logps/chosen": -0.9382683634757996, | |
| "logps/rejected": -1.1057795286178589, | |
| "loss": 1.14, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -1.8765367269515991, | |
| "rewards/margins": 0.3350227475166321, | |
| "rewards/rejected": -2.2115590572357178, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 2.0677233429394812, | |
| "grad_norm": 17.33211536858926, | |
| "learning_rate": 1.3325565235427716e-08, | |
| "logits/chosen": -2.028552770614624, | |
| "logits/rejected": -2.0268807411193848, | |
| "logps/chosen": -0.9831819534301758, | |
| "logps/rejected": -1.1274645328521729, | |
| "loss": 1.177, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9663639068603516, | |
| "rewards/margins": 0.288565069437027, | |
| "rewards/rejected": -2.2549290657043457, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 2.07492795389049, | |
| "grad_norm": 17.173051243263835, | |
| "learning_rate": 1.3140628540053218e-08, | |
| "logits/chosen": -1.9946448802947998, | |
| "logits/rejected": -1.9966709613800049, | |
| "logps/chosen": -0.9759955406188965, | |
| "logps/rejected": -1.1114940643310547, | |
| "loss": 1.1833, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.951991081237793, | |
| "rewards/margins": 0.2709970772266388, | |
| "rewards/rejected": -2.2229881286621094, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 2.0821325648414986, | |
| "grad_norm": 19.25779046293631, | |
| "learning_rate": 1.2956525509522451e-08, | |
| "logits/chosen": -1.9791204929351807, | |
| "logits/rejected": -1.97879159450531, | |
| "logps/chosen": -1.1120542287826538, | |
| "logps/rejected": -1.2156860828399658, | |
| "loss": 1.234, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.2241084575653076, | |
| "rewards/margins": 0.20726370811462402, | |
| "rewards/rejected": -2.4313721656799316, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 2.089337175792507, | |
| "grad_norm": 19.696970893217582, | |
| "learning_rate": 1.2773269085518267e-08, | |
| "logits/chosen": -2.011164426803589, | |
| "logits/rejected": -2.0127670764923096, | |
| "logps/chosen": -1.0766938924789429, | |
| "logps/rejected": -1.2073553800582886, | |
| "loss": 1.1837, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -2.1533877849578857, | |
| "rewards/margins": 0.2613227963447571, | |
| "rewards/rejected": -2.414710760116577, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.096541786743516, | |
| "grad_norm": 20.176935063380885, | |
| "learning_rate": 1.2590872150210574e-08, | |
| "logits/chosen": -2.0675017833709717, | |
| "logits/rejected": -2.0605273246765137, | |
| "logps/chosen": -1.057830810546875, | |
| "logps/rejected": -1.1687963008880615, | |
| "loss": 1.2247, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.11566162109375, | |
| "rewards/margins": 0.22193074226379395, | |
| "rewards/rejected": -2.337592601776123, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 2.1037463976945245, | |
| "grad_norm": 20.572020361191345, | |
| "learning_rate": 1.2409347525350775e-08, | |
| "logits/chosen": -2.0273613929748535, | |
| "logits/rejected": -2.0172836780548096, | |
| "logps/chosen": -1.1093732118606567, | |
| "logps/rejected": -1.2572507858276367, | |
| "loss": 1.166, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.2187464237213135, | |
| "rewards/margins": 0.2957550585269928, | |
| "rewards/rejected": -2.5145015716552734, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 2.110951008645533, | |
| "grad_norm": 22.752346590850024, | |
| "learning_rate": 1.2228707971370421e-08, | |
| "logits/chosen": -2.018433094024658, | |
| "logits/rejected": -2.01145601272583, | |
| "logps/chosen": -0.9928643107414246, | |
| "logps/rejected": -1.1063158512115479, | |
| "loss": 1.2256, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.9857286214828491, | |
| "rewards/margins": 0.22690317034721375, | |
| "rewards/rejected": -2.2126317024230957, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 2.118155619596542, | |
| "grad_norm": 21.233577131341413, | |
| "learning_rate": 1.2048966186484282e-08, | |
| "logits/chosen": -2.015031337738037, | |
| "logits/rejected": -1.9982612133026123, | |
| "logps/chosen": -1.117865800857544, | |
| "logps/rejected": -1.232062578201294, | |
| "loss": 1.2127, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.235731601715088, | |
| "rewards/margins": 0.2283933460712433, | |
| "rewards/rejected": -2.464125156402588, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 2.1253602305475505, | |
| "grad_norm": 28.100094599633593, | |
| "learning_rate": 1.187013480579762e-08, | |
| "logits/chosen": -2.010659694671631, | |
| "logits/rejected": -2.0132699012756348, | |
| "logps/chosen": -1.0425684452056885, | |
| "logps/rejected": -1.1762912273406982, | |
| "loss": 1.201, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.085136890411377, | |
| "rewards/margins": 0.26744550466537476, | |
| "rewards/rejected": -2.3525824546813965, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.132564841498559, | |
| "grad_norm": 40.79478390193748, | |
| "learning_rate": 1.1692226400418073e-08, | |
| "logits/chosen": -1.9510002136230469, | |
| "logits/rejected": -1.9495048522949219, | |
| "logps/chosen": -1.0822184085845947, | |
| "logps/rejected": -1.212081789970398, | |
| "loss": 1.2188, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.1644368171691895, | |
| "rewards/margins": 0.2597268521785736, | |
| "rewards/rejected": -2.424163579940796, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 2.139769452449568, | |
| "grad_norm": 16.326675723252357, | |
| "learning_rate": 1.1515253476571923e-08, | |
| "logits/chosen": -1.9815658330917358, | |
| "logits/rejected": -1.975783109664917, | |
| "logps/chosen": -1.0105996131896973, | |
| "logps/rejected": -1.192287564277649, | |
| "loss": 1.1208, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.0211992263793945, | |
| "rewards/margins": 0.36337584257125854, | |
| "rewards/rejected": -2.384575128555298, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 2.1469740634005765, | |
| "grad_norm": 19.921977505309442, | |
| "learning_rate": 1.133922847472496e-08, | |
| "logits/chosen": -1.9953126907348633, | |
| "logits/rejected": -1.9960988759994507, | |
| "logps/chosen": -1.110705018043518, | |
| "logps/rejected": -1.209160566329956, | |
| "loss": 1.2518, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.221410036087036, | |
| "rewards/margins": 0.19691102206707, | |
| "rewards/rejected": -2.418321132659912, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 2.154178674351585, | |
| "grad_norm": 22.89240067306987, | |
| "learning_rate": 1.1164163768707952e-08, | |
| "logits/chosen": -2.003279209136963, | |
| "logits/rejected": -1.9980405569076538, | |
| "logps/chosen": -1.0043981075286865, | |
| "logps/rejected": -1.1427768468856812, | |
| "loss": 1.1843, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.008796215057373, | |
| "rewards/margins": 0.27675721049308777, | |
| "rewards/rejected": -2.2855536937713623, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 2.161383285302594, | |
| "grad_norm": 17.89862895130365, | |
| "learning_rate": 1.0990071664846861e-08, | |
| "logits/chosen": -1.9780519008636475, | |
| "logits/rejected": -1.9771487712860107, | |
| "logps/chosen": -1.0197398662567139, | |
| "logps/rejected": -1.1996923685073853, | |
| "loss": 1.1581, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0394797325134277, | |
| "rewards/margins": 0.35990482568740845, | |
| "rewards/rejected": -2.3993847370147705, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.1685878962536025, | |
| "grad_norm": 18.90659171579793, | |
| "learning_rate": 1.0816964401097739e-08, | |
| "logits/chosen": -1.964535117149353, | |
| "logits/rejected": -1.9613316059112549, | |
| "logps/chosen": -0.9563964009284973, | |
| "logps/rejected": -1.0800330638885498, | |
| "loss": 1.2052, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -1.9127928018569946, | |
| "rewards/margins": 0.24727335572242737, | |
| "rewards/rejected": -2.1600661277770996, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 2.175792507204611, | |
| "grad_norm": 19.878484764331017, | |
| "learning_rate": 1.0644854146186406e-08, | |
| "logits/chosen": -2.0236928462982178, | |
| "logits/rejected": -2.0173866748809814, | |
| "logps/chosen": -1.0241403579711914, | |
| "logps/rejected": -1.1835166215896606, | |
| "loss": 1.1624, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.048280715942383, | |
| "rewards/margins": 0.318752646446228, | |
| "rewards/rejected": -2.3670332431793213, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 2.18299711815562, | |
| "grad_norm": 19.088076538610206, | |
| "learning_rate": 1.0473752998753114e-08, | |
| "logits/chosen": -2.004102945327759, | |
| "logits/rejected": -1.9954335689544678, | |
| "logps/chosen": -1.0195882320404053, | |
| "logps/rejected": -1.1798489093780518, | |
| "loss": 1.1535, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0391764640808105, | |
| "rewards/margins": 0.3205214738845825, | |
| "rewards/rejected": -2.3596978187561035, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 2.1902017291066285, | |
| "grad_norm": 19.31361091042759, | |
| "learning_rate": 1.030367298650201e-08, | |
| "logits/chosen": -2.023881196975708, | |
| "logits/rejected": -2.0238354206085205, | |
| "logps/chosen": -1.0392138957977295, | |
| "logps/rejected": -1.19191312789917, | |
| "loss": 1.1578, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -2.078427791595459, | |
| "rewards/margins": 0.3053986132144928, | |
| "rewards/rejected": -2.38382625579834, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 2.1974063400576367, | |
| "grad_norm": 22.093759972479646, | |
| "learning_rate": 1.0134626065355675e-08, | |
| "logits/chosen": -2.0746548175811768, | |
| "logits/rejected": -2.0715177059173584, | |
| "logps/chosen": -1.0234037637710571, | |
| "logps/rejected": -1.1665077209472656, | |
| "loss": 1.1883, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.0468075275421143, | |
| "rewards/margins": 0.2862081527709961, | |
| "rewards/rejected": -2.3330154418945312, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 2.2046109510086453, | |
| "grad_norm": 19.64286406855496, | |
| "learning_rate": 9.966624118614611e-09, | |
| "logits/chosen": -2.006706476211548, | |
| "logits/rejected": -2.0016489028930664, | |
| "logps/chosen": -1.0631506443023682, | |
| "logps/rejected": -1.2085468769073486, | |
| "loss": 1.1876, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.1263012886047363, | |
| "rewards/margins": 0.29079198837280273, | |
| "rewards/rejected": -2.4170937538146973, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 2.211815561959654, | |
| "grad_norm": 14.856155733229528, | |
| "learning_rate": 9.799678956121976e-09, | |
| "logits/chosen": -1.9717843532562256, | |
| "logits/rejected": -1.9674240350723267, | |
| "logps/chosen": -1.0307656526565552, | |
| "logps/rejected": -1.1394712924957275, | |
| "loss": 1.2006, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.0615313053131104, | |
| "rewards/margins": 0.21741144359111786, | |
| "rewards/rejected": -2.278942584991455, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 2.2190201729106627, | |
| "grad_norm": 23.633018389781732, | |
| "learning_rate": 9.633802313433314e-09, | |
| "logits/chosen": -1.9454095363616943, | |
| "logits/rejected": -1.9511306285858154, | |
| "logps/chosen": -1.0190519094467163, | |
| "logps/rejected": -1.1248835325241089, | |
| "loss": 1.2055, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.0381038188934326, | |
| "rewards/margins": 0.21166305243968964, | |
| "rewards/rejected": -2.2497670650482178, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 2.2262247838616713, | |
| "grad_norm": 20.794315619142072, | |
| "learning_rate": 9.469005850991705e-09, | |
| "logits/chosen": -2.0088305473327637, | |
| "logits/rejected": -2.003154754638672, | |
| "logps/chosen": -1.0144597291946411, | |
| "logps/rejected": -1.1316652297973633, | |
| "loss": 1.2343, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.0289194583892822, | |
| "rewards/margins": 0.23441116511821747, | |
| "rewards/rejected": -2.2633304595947266, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 2.23342939481268, | |
| "grad_norm": 18.67038535819961, | |
| "learning_rate": 9.305301153307949e-09, | |
| "logits/chosen": -2.0057482719421387, | |
| "logits/rejected": -2.0133614540100098, | |
| "logps/chosen": -0.9462668299674988, | |
| "logps/rejected": -1.1108109951019287, | |
| "loss": 1.1573, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -1.8925336599349976, | |
| "rewards/margins": 0.3290883004665375, | |
| "rewards/rejected": -2.2216219902038574, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.2406340057636887, | |
| "grad_norm": 18.08187045245269, | |
| "learning_rate": 9.142699728146336e-09, | |
| "logits/chosen": -1.9763425588607788, | |
| "logits/rejected": -1.9695403575897217, | |
| "logps/chosen": -1.0319360494613647, | |
| "logps/rejected": -1.1644192934036255, | |
| "loss": 1.2014, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.0638720989227295, | |
| "rewards/margins": 0.26496636867523193, | |
| "rewards/rejected": -2.328838586807251, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 2.2478386167146973, | |
| "grad_norm": 16.765059853307356, | |
| "learning_rate": 8.981213005715627e-09, | |
| "logits/chosen": -2.002300500869751, | |
| "logits/rejected": -2.005335569381714, | |
| "logps/chosen": -0.9925374984741211, | |
| "logps/rejected": -1.1654067039489746, | |
| "loss": 1.1484, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -1.9850749969482422, | |
| "rewards/margins": 0.34573858976364136, | |
| "rewards/rejected": -2.330813407897949, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 2.255043227665706, | |
| "grad_norm": 21.826707648017194, | |
| "learning_rate": 8.820852337865611e-09, | |
| "logits/chosen": -2.0354738235473633, | |
| "logits/rejected": -2.031705617904663, | |
| "logps/chosen": -0.9956668019294739, | |
| "logps/rejected": -1.1440733671188354, | |
| "loss": 1.1717, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -1.9913336038589478, | |
| "rewards/margins": 0.296813428401947, | |
| "rewards/rejected": -2.288146734237671, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 2.2622478386167146, | |
| "grad_norm": 17.005533531498173, | |
| "learning_rate": 8.661628997289044e-09, | |
| "logits/chosen": -1.9752384424209595, | |
| "logits/rejected": -1.97113835811615, | |
| "logps/chosen": -1.0153406858444214, | |
| "logps/rejected": -1.1710517406463623, | |
| "loss": 1.1687, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0306813716888428, | |
| "rewards/margins": 0.3114221394062042, | |
| "rewards/rejected": -2.3421034812927246, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 2.2694524495677233, | |
| "grad_norm": 16.411029038337308, | |
| "learning_rate": 8.503554176729341e-09, | |
| "logits/chosen": -1.976362943649292, | |
| "logits/rejected": -1.974590539932251, | |
| "logps/chosen": -1.026755928993225, | |
| "logps/rejected": -1.1865880489349365, | |
| "loss": 1.1696, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.05351185798645, | |
| "rewards/margins": 0.3196641802787781, | |
| "rewards/rejected": -2.373176097869873, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.276657060518732, | |
| "grad_norm": 24.940392400474, | |
| "learning_rate": 8.346638988193636e-09, | |
| "logits/chosen": -2.0030248165130615, | |
| "logits/rejected": -1.9979517459869385, | |
| "logps/chosen": -0.9251815676689148, | |
| "logps/rejected": -1.0761079788208008, | |
| "loss": 1.1768, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -1.8503631353378296, | |
| "rewards/margins": 0.30185258388519287, | |
| "rewards/rejected": -2.1522159576416016, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 2.2838616714697406, | |
| "grad_norm": 23.2363909978899, | |
| "learning_rate": 8.19089446217176e-09, | |
| "logits/chosen": -1.9777719974517822, | |
| "logits/rejected": -1.9676783084869385, | |
| "logps/chosen": -1.0022261142730713, | |
| "logps/rejected": -1.1919556856155396, | |
| "loss": 1.1202, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -2.0044522285461426, | |
| "rewards/margins": 0.3794591426849365, | |
| "rewards/rejected": -2.383911371231079, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 2.2910662824207493, | |
| "grad_norm": 17.14618373707155, | |
| "learning_rate": 8.036331546860777e-09, | |
| "logits/chosen": -1.982940435409546, | |
| "logits/rejected": -1.982465386390686, | |
| "logps/chosen": -0.9494163393974304, | |
| "logps/rejected": -1.039945125579834, | |
| "loss": 1.2442, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -1.8988326787948608, | |
| "rewards/margins": 0.18105748295783997, | |
| "rewards/rejected": -2.079890251159668, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 2.298270893371758, | |
| "grad_norm": 23.5198522631464, | |
| "learning_rate": 7.882961107395416e-09, | |
| "logits/chosen": -1.9984643459320068, | |
| "logits/rejected": -1.9926011562347412, | |
| "logps/chosen": -1.130748987197876, | |
| "logps/rejected": -1.177819848060608, | |
| "loss": 1.315, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -2.261497974395752, | |
| "rewards/margins": 0.09414196014404297, | |
| "rewards/rejected": -2.355639696121216, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 2.3054755043227666, | |
| "grad_norm": 25.590315233089598, | |
| "learning_rate": 7.73079392508428e-09, | |
| "logits/chosen": -1.9712879657745361, | |
| "logits/rejected": -1.9705880880355835, | |
| "logps/chosen": -1.0907241106033325, | |
| "logps/rejected": -1.2799861431121826, | |
| "loss": 1.1523, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.181448221206665, | |
| "rewards/margins": 0.37852445244789124, | |
| "rewards/rejected": -2.5599722862243652, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.3126801152737753, | |
| "grad_norm": 21.478168268234054, | |
| "learning_rate": 7.579840696651938e-09, | |
| "logits/chosen": -1.9963871240615845, | |
| "logits/rejected": -1.9932276010513306, | |
| "logps/chosen": -1.0474956035614014, | |
| "logps/rejected": -1.172515869140625, | |
| "loss": 1.2079, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0949912071228027, | |
| "rewards/margins": 0.2500404119491577, | |
| "rewards/rejected": -2.34503173828125, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 2.319884726224784, | |
| "grad_norm": 20.9972814315902, | |
| "learning_rate": 7.43011203348704e-09, | |
| "logits/chosen": -1.9149713516235352, | |
| "logits/rejected": -1.9116861820220947, | |
| "logps/chosen": -1.0514217615127563, | |
| "logps/rejected": -1.1269280910491943, | |
| "loss": 1.2686, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.1028435230255127, | |
| "rewards/margins": 0.1510128229856491, | |
| "rewards/rejected": -2.2538561820983887, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 2.3270893371757926, | |
| "grad_norm": 18.760848272652197, | |
| "learning_rate": 7.281618460896344e-09, | |
| "logits/chosen": -1.995486855506897, | |
| "logits/rejected": -1.9929373264312744, | |
| "logps/chosen": -0.9654563665390015, | |
| "logps/rejected": -1.1074378490447998, | |
| "loss": 1.1729, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.930912733078003, | |
| "rewards/margins": 0.2839628756046295, | |
| "rewards/rejected": -2.2148756980895996, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 2.3342939481268012, | |
| "grad_norm": 20.42845258559301, | |
| "learning_rate": 7.134370417364849e-09, | |
| "logits/chosen": -1.9683917760849, | |
| "logits/rejected": -1.9679603576660156, | |
| "logps/chosen": -1.001461386680603, | |
| "logps/rejected": -1.1403329372406006, | |
| "loss": 1.2027, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.002922773361206, | |
| "rewards/margins": 0.2777433395385742, | |
| "rewards/rejected": -2.280665874481201, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 2.34149855907781, | |
| "grad_norm": 23.164358986342677, | |
| "learning_rate": 6.988378253821981e-09, | |
| "logits/chosen": -1.9697679281234741, | |
| "logits/rejected": -1.9687258005142212, | |
| "logps/chosen": -1.0258748531341553, | |
| "logps/rejected": -1.143920660018921, | |
| "loss": 1.209, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0517497062683105, | |
| "rewards/margins": 0.23609168827533722, | |
| "rewards/rejected": -2.287841320037842, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.3487031700288186, | |
| "grad_norm": 20.30991552682094, | |
| "learning_rate": 6.8436522329140186e-09, | |
| "logits/chosen": -1.9788017272949219, | |
| "logits/rejected": -1.985569715499878, | |
| "logps/chosen": -1.0339092016220093, | |
| "logps/rejected": -1.1592271327972412, | |
| "loss": 1.2106, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.0678184032440186, | |
| "rewards/margins": 0.2506362795829773, | |
| "rewards/rejected": -2.3184542655944824, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 2.3559077809798272, | |
| "grad_norm": 21.894995604840652, | |
| "learning_rate": 6.700202528282603e-09, | |
| "logits/chosen": -1.978734016418457, | |
| "logits/rejected": -1.969061255455017, | |
| "logps/chosen": -1.0286333560943604, | |
| "logps/rejected": -1.1439108848571777, | |
| "loss": 1.2158, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0572667121887207, | |
| "rewards/margins": 0.23055517673492432, | |
| "rewards/rejected": -2.2878217697143555, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 2.363112391930836, | |
| "grad_norm": 21.434677454334327, | |
| "learning_rate": 6.558039223849668e-09, | |
| "logits/chosen": -2.0271782875061035, | |
| "logits/rejected": -2.0174622535705566, | |
| "logps/chosen": -1.0365641117095947, | |
| "logps/rejected": -1.2443287372589111, | |
| "loss": 1.1155, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.0731282234191895, | |
| "rewards/margins": 0.41552942991256714, | |
| "rewards/rejected": -2.4886574745178223, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 2.3703170028818445, | |
| "grad_norm": 22.134638764373964, | |
| "learning_rate": 6.417172313108471e-09, | |
| "logits/chosen": -1.95876944065094, | |
| "logits/rejected": -1.9533073902130127, | |
| "logps/chosen": -0.9859912991523743, | |
| "logps/rejected": -1.1151400804519653, | |
| "loss": 1.1979, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9719825983047485, | |
| "rewards/margins": 0.2582974135875702, | |
| "rewards/rejected": -2.2302801609039307, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 2.377521613832853, | |
| "grad_norm": 21.658570611710445, | |
| "learning_rate": 6.277611698421179e-09, | |
| "logits/chosen": -2.017606735229492, | |
| "logits/rejected": -2.0095458030700684, | |
| "logps/chosen": -0.9050453305244446, | |
| "logps/rejected": -1.098288893699646, | |
| "loss": 1.1252, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -1.8100906610488892, | |
| "rewards/margins": 0.3864876627922058, | |
| "rewards/rejected": -2.196577787399292, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.3847262247838614, | |
| "grad_norm": 22.8559892529762, | |
| "learning_rate": 6.139367190322714e-09, | |
| "logits/chosen": -2.0034892559051514, | |
| "logits/rejected": -2.0032081604003906, | |
| "logps/chosen": -1.0592529773712158, | |
| "logps/rejected": -1.2185790538787842, | |
| "loss": 1.1609, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.1185059547424316, | |
| "rewards/margins": 0.3186524510383606, | |
| "rewards/rejected": -2.4371581077575684, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 2.39193083573487, | |
| "grad_norm": 17.198608533100995, | |
| "learning_rate": 6.002448506831171e-09, | |
| "logits/chosen": -2.0061838626861572, | |
| "logits/rejected": -2.0014090538024902, | |
| "logps/chosen": -0.9808699488639832, | |
| "logps/rejected": -1.1246802806854248, | |
| "loss": 1.1731, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.9617398977279663, | |
| "rewards/margins": 0.2876203954219818, | |
| "rewards/rejected": -2.2493605613708496, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 2.3991354466858787, | |
| "grad_norm": 18.199025209277288, | |
| "learning_rate": 5.866865272764607e-09, | |
| "logits/chosen": -2.023648262023926, | |
| "logits/rejected": -2.0234923362731934, | |
| "logps/chosen": -1.0167878866195679, | |
| "logps/rejected": -1.1612762212753296, | |
| "loss": 1.1772, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.0335757732391357, | |
| "rewards/margins": 0.28897663950920105, | |
| "rewards/rejected": -2.322552442550659, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 2.4063400576368874, | |
| "grad_norm": 23.302252487813124, | |
| "learning_rate": 5.7326270190645595e-09, | |
| "logits/chosen": -1.896691918373108, | |
| "logits/rejected": -1.8979320526123047, | |
| "logps/chosen": -1.0594362020492554, | |
| "logps/rejected": -1.1698405742645264, | |
| "loss": 1.2168, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.1188724040985107, | |
| "rewards/margins": 0.2208089381456375, | |
| "rewards/rejected": -2.3396811485290527, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 2.413544668587896, | |
| "grad_norm": 18.446092862588884, | |
| "learning_rate": 5.599743182125938e-09, | |
| "logits/chosen": -2.043023109436035, | |
| "logits/rejected": -2.043013095855713, | |
| "logps/chosen": -1.0480068922042847, | |
| "logps/rejected": -1.1850215196609497, | |
| "loss": 1.179, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.0960137844085693, | |
| "rewards/margins": 0.2740294933319092, | |
| "rewards/rejected": -2.3700430393218994, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.4207492795389047, | |
| "grad_norm": 20.220307143059344, | |
| "learning_rate": 5.46822310313379e-09, | |
| "logits/chosen": -2.0473732948303223, | |
| "logits/rejected": -2.0569522380828857, | |
| "logps/chosen": -1.09086012840271, | |
| "logps/rejected": -1.1955832242965698, | |
| "loss": 1.2353, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.18172025680542, | |
| "rewards/margins": 0.2094462662935257, | |
| "rewards/rejected": -2.3911664485931396, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 2.4279538904899134, | |
| "grad_norm": 20.780990431383444, | |
| "learning_rate": 5.33807602740658e-09, | |
| "logits/chosen": -2.022789478302002, | |
| "logits/rejected": -2.0159573554992676, | |
| "logps/chosen": -0.9560559988021851, | |
| "logps/rejected": -1.160628318786621, | |
| "loss": 1.111, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -1.9121119976043701, | |
| "rewards/margins": 0.40914446115493774, | |
| "rewards/rejected": -2.321256637573242, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 2.435158501440922, | |
| "grad_norm": 21.245348975655457, | |
| "learning_rate": 5.209311103746334e-09, | |
| "logits/chosen": -2.0008084774017334, | |
| "logits/rejected": -2.0011303424835205, | |
| "logps/chosen": -1.0523884296417236, | |
| "logps/rejected": -1.224974274635315, | |
| "loss": 1.1587, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.1047768592834473, | |
| "rewards/margins": 0.3451715409755707, | |
| "rewards/rejected": -2.44994854927063, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 2.4423631123919307, | |
| "grad_norm": 24.352598699910715, | |
| "learning_rate": 5.081937383795484e-09, | |
| "logits/chosen": -1.972608208656311, | |
| "logits/rejected": -1.9717302322387695, | |
| "logps/chosen": -0.9721845388412476, | |
| "logps/rejected": -1.137112021446228, | |
| "loss": 1.1484, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -1.9443690776824951, | |
| "rewards/margins": 0.32985490560531616, | |
| "rewards/rejected": -2.274224042892456, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 2.4495677233429394, | |
| "grad_norm": 18.574622449743107, | |
| "learning_rate": 4.955963821400599e-09, | |
| "logits/chosen": -2.0249781608581543, | |
| "logits/rejected": -2.019134759902954, | |
| "logps/chosen": -1.029394507408142, | |
| "logps/rejected": -1.167999505996704, | |
| "loss": 1.1922, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.058789014816284, | |
| "rewards/margins": 0.2772100567817688, | |
| "rewards/rejected": -2.335999011993408, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.456772334293948, | |
| "grad_norm": 15.429482416255146, | |
| "learning_rate": 4.831399271982928e-09, | |
| "logits/chosen": -1.9512850046157837, | |
| "logits/rejected": -1.9432300329208374, | |
| "logps/chosen": -1.0431114435195923, | |
| "logps/rejected": -1.1738238334655762, | |
| "loss": 1.2095, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.0862228870391846, | |
| "rewards/margins": 0.26142507791519165, | |
| "rewards/rejected": -2.3476476669311523, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 2.4639769452449567, | |
| "grad_norm": 25.103110732614255, | |
| "learning_rate": 4.708252491915951e-09, | |
| "logits/chosen": -2.030299663543701, | |
| "logits/rejected": -2.024203300476074, | |
| "logps/chosen": -1.0459200143814087, | |
| "logps/rejected": -1.1930789947509766, | |
| "loss": 1.197, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0918400287628174, | |
| "rewards/margins": 0.2943178713321686, | |
| "rewards/rejected": -2.386157989501953, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 2.4711815561959654, | |
| "grad_norm": 25.280433628761124, | |
| "learning_rate": 4.58653213790981e-09, | |
| "logits/chosen": -2.006598472595215, | |
| "logits/rejected": -1.9983857870101929, | |
| "logps/chosen": -1.025721549987793, | |
| "logps/rejected": -1.1746145486831665, | |
| "loss": 1.1792, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.051443099975586, | |
| "rewards/margins": 0.29778599739074707, | |
| "rewards/rejected": -2.349229097366333, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 2.478386167146974, | |
| "grad_norm": 18.242083284353217, | |
| "learning_rate": 4.466246766402773e-09, | |
| "logits/chosen": -1.9907543659210205, | |
| "logits/rejected": -1.9845359325408936, | |
| "logps/chosen": -1.0393613576889038, | |
| "logps/rejected": -1.1935051679611206, | |
| "loss": 1.1827, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0787227153778076, | |
| "rewards/margins": 0.30828770995140076, | |
| "rewards/rejected": -2.387010335922241, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 2.4855907780979827, | |
| "grad_norm": 22.018503196573274, | |
| "learning_rate": 4.347404832959775e-09, | |
| "logits/chosen": -2.0336432456970215, | |
| "logits/rejected": -2.0338807106018066, | |
| "logps/chosen": -1.032832384109497, | |
| "logps/rejected": -1.1933454275131226, | |
| "loss": 1.1612, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.065664768218994, | |
| "rewards/margins": 0.32102587819099426, | |
| "rewards/rejected": -2.386690855026245, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.4927953890489913, | |
| "grad_norm": 33.037428693429234, | |
| "learning_rate": 4.230014691678016e-09, | |
| "logits/chosen": -1.9883922338485718, | |
| "logits/rejected": -1.9890626668930054, | |
| "logps/chosen": -1.0595102310180664, | |
| "logps/rejected": -1.126479148864746, | |
| "loss": 1.272, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.119020462036133, | |
| "rewards/margins": 0.1339379847049713, | |
| "rewards/rejected": -2.252958297729492, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 17.82843912451702, | |
| "learning_rate": 4.114084594599707e-09, | |
| "logits/chosen": -1.9903564453125, | |
| "logits/rejected": -1.9900470972061157, | |
| "logps/chosen": -1.0114375352859497, | |
| "logps/rejected": -1.229552984237671, | |
| "loss": 1.1011, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -2.0228750705718994, | |
| "rewards/margins": 0.4362305998802185, | |
| "rewards/rejected": -2.459105968475342, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 2.5072046109510087, | |
| "grad_norm": 22.102059612075095, | |
| "learning_rate": 3.9996226911319546e-09, | |
| "logits/chosen": -1.993326187133789, | |
| "logits/rejected": -1.981066107749939, | |
| "logps/chosen": -1.0159366130828857, | |
| "logps/rejected": -1.1457350254058838, | |
| "loss": 1.1898, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.0318732261657715, | |
| "rewards/margins": 0.25959664583206177, | |
| "rewards/rejected": -2.2914700508117676, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 2.5144092219020173, | |
| "grad_norm": 17.973603590541654, | |
| "learning_rate": 3.886637027473949e-09, | |
| "logits/chosen": -2.0013790130615234, | |
| "logits/rejected": -2.0035085678100586, | |
| "logps/chosen": -1.076293706893921, | |
| "logps/rejected": -1.2393258810043335, | |
| "loss": 1.1562, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.152587413787842, | |
| "rewards/margins": 0.3260645270347595, | |
| "rewards/rejected": -2.478651762008667, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 2.521613832853026, | |
| "grad_norm": 19.37957776631117, | |
| "learning_rate": 3.775135546051295e-09, | |
| "logits/chosen": -1.9389715194702148, | |
| "logits/rejected": -1.9399712085723877, | |
| "logps/chosen": -1.0256609916687012, | |
| "logps/rejected": -1.1517935991287231, | |
| "loss": 1.1987, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0513219833374023, | |
| "rewards/margins": 0.252265065908432, | |
| "rewards/rejected": -2.3035871982574463, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.5288184438040346, | |
| "grad_norm": 23.33567215234884, | |
| "learning_rate": 3.665126084957723e-09, | |
| "logits/chosen": -1.984487533569336, | |
| "logits/rejected": -1.9886258840560913, | |
| "logps/chosen": -1.1339917182922363, | |
| "logps/rejected": -1.233039140701294, | |
| "loss": 1.2585, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.2679834365844727, | |
| "rewards/margins": 0.19809459149837494, | |
| "rewards/rejected": -2.466078281402588, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 2.5360230547550433, | |
| "grad_norm": 19.609640038869685, | |
| "learning_rate": 3.556616377404101e-09, | |
| "logits/chosen": -2.00850248336792, | |
| "logits/rejected": -2.006412982940674, | |
| "logps/chosen": -1.07861328125, | |
| "logps/rejected": -1.236485242843628, | |
| "loss": 1.154, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.1572265625, | |
| "rewards/margins": 0.31574416160583496, | |
| "rewards/rejected": -2.472970485687256, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 2.543227665706052, | |
| "grad_norm": 19.767394228725337, | |
| "learning_rate": 3.4496140511748125e-09, | |
| "logits/chosen": -1.9994624853134155, | |
| "logits/rejected": -1.9942439794540405, | |
| "logps/chosen": -1.0551049709320068, | |
| "logps/rejected": -1.1994330883026123, | |
| "loss": 1.1752, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.1102099418640137, | |
| "rewards/margins": 0.28865596652030945, | |
| "rewards/rejected": -2.3988661766052246, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 2.5504322766570606, | |
| "grad_norm": 31.22852578343729, | |
| "learning_rate": 3.3441266280915427e-09, | |
| "logits/chosen": -1.9827390909194946, | |
| "logits/rejected": -1.983473539352417, | |
| "logps/chosen": -1.093752145767212, | |
| "logps/rejected": -1.20872163772583, | |
| "loss": 1.2131, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.187504291534424, | |
| "rewards/margins": 0.229939267039299, | |
| "rewards/rejected": -2.41744327545166, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 2.5576368876080693, | |
| "grad_norm": 23.706412362537016, | |
| "learning_rate": 3.2401615234845693e-09, | |
| "logits/chosen": -2.0068211555480957, | |
| "logits/rejected": -2.0008292198181152, | |
| "logps/chosen": -1.0923867225646973, | |
| "logps/rejected": -1.2357128858566284, | |
| "loss": 1.1896, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1847734451293945, | |
| "rewards/margins": 0.28665226697921753, | |
| "rewards/rejected": -2.471425771713257, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.564841498559078, | |
| "grad_norm": 16.13728529223842, | |
| "learning_rate": 3.1377260456714375e-09, | |
| "logits/chosen": -1.901414155960083, | |
| "logits/rejected": -1.8929615020751953, | |
| "logps/chosen": -1.0596764087677002, | |
| "logps/rejected": -1.2023025751113892, | |
| "loss": 1.1686, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.1193528175354004, | |
| "rewards/margins": 0.28525251150131226, | |
| "rewards/rejected": -2.4046051502227783, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 2.5720461095100866, | |
| "grad_norm": 18.028717215705484, | |
| "learning_rate": 3.0368273954432698e-09, | |
| "logits/chosen": -2.0307531356811523, | |
| "logits/rejected": -2.022324800491333, | |
| "logps/chosen": -1.0493528842926025, | |
| "logps/rejected": -1.1534329652786255, | |
| "loss": 1.2252, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.098705768585205, | |
| "rewards/margins": 0.20816004276275635, | |
| "rewards/rejected": -2.306865930557251, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 2.5792507204610953, | |
| "grad_norm": 17.367490930434325, | |
| "learning_rate": 2.937472665558541e-09, | |
| "logits/chosen": -2.019484281539917, | |
| "logits/rejected": -2.020643711090088, | |
| "logps/chosen": -1.036195993423462, | |
| "logps/rejected": -1.147991418838501, | |
| "loss": 1.2267, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.072391986846924, | |
| "rewards/margins": 0.22359101474285126, | |
| "rewards/rejected": -2.295982837677002, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 2.586455331412104, | |
| "grad_norm": 21.805325598847563, | |
| "learning_rate": 2.8396688402445053e-09, | |
| "logits/chosen": -2.0637335777282715, | |
| "logits/rejected": -2.0563552379608154, | |
| "logps/chosen": -1.0100147724151611, | |
| "logps/rejected": -1.2180942296981812, | |
| "loss": 1.1063, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -2.0200295448303223, | |
| "rewards/margins": 0.4161592125892639, | |
| "rewards/rejected": -2.4361884593963623, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 2.5936599423631126, | |
| "grad_norm": 24.439180591540023, | |
| "learning_rate": 2.7434227947062324e-09, | |
| "logits/chosen": -2.0062127113342285, | |
| "logits/rejected": -1.9998852014541626, | |
| "logps/chosen": -1.1317923069000244, | |
| "logps/rejected": -1.2398041486740112, | |
| "loss": 1.2326, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.263584613800049, | |
| "rewards/margins": 0.21602365374565125, | |
| "rewards/rejected": -2.4796082973480225, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.6008645533141213, | |
| "grad_norm": 18.265831934479376, | |
| "learning_rate": 2.6487412946432976e-09, | |
| "logits/chosen": -1.9716873168945312, | |
| "logits/rejected": -1.966560959815979, | |
| "logps/chosen": -1.0693469047546387, | |
| "logps/rejected": -1.2056225538253784, | |
| "loss": 1.1912, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.1386938095092773, | |
| "rewards/margins": 0.27255168557167053, | |
| "rewards/rejected": -2.411245107650757, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 2.60806916426513, | |
| "grad_norm": 22.910160520824302, | |
| "learning_rate": 2.5556309957742024e-09, | |
| "logits/chosen": -1.97675359249115, | |
| "logits/rejected": -1.9716304540634155, | |
| "logps/chosen": -1.0250674486160278, | |
| "logps/rejected": -1.2212371826171875, | |
| "loss": 1.1161, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -2.0501348972320557, | |
| "rewards/margins": 0.3923397660255432, | |
| "rewards/rejected": -2.442474365234375, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 2.6152737752161386, | |
| "grad_norm": 22.671601957903725, | |
| "learning_rate": 2.4640984433684758e-09, | |
| "logits/chosen": -2.0380005836486816, | |
| "logits/rejected": -2.0387332439422607, | |
| "logps/chosen": -1.1190853118896484, | |
| "logps/rejected": -1.235012173652649, | |
| "loss": 1.2353, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -2.238170623779297, | |
| "rewards/margins": 0.231853649020195, | |
| "rewards/rejected": -2.470024347305298, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 2.6224783861671472, | |
| "grad_norm": 17.024567886757257, | |
| "learning_rate": 2.3741500717865987e-09, | |
| "logits/chosen": -1.9916216135025024, | |
| "logits/rejected": -2.0025291442871094, | |
| "logps/chosen": -1.0068811178207397, | |
| "logps/rejected": -1.1515626907348633, | |
| "loss": 1.1789, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.0137622356414795, | |
| "rewards/margins": 0.2893627882003784, | |
| "rewards/rejected": -2.3031253814697266, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 2.629682997118156, | |
| "grad_norm": 17.494575910236158, | |
| "learning_rate": 2.285792204027678e-09, | |
| "logits/chosen": -1.9781382083892822, | |
| "logits/rejected": -1.9753141403198242, | |
| "logps/chosen": -1.013346791267395, | |
| "logps/rejected": -1.211428165435791, | |
| "loss": 1.1021, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": -2.02669358253479, | |
| "rewards/margins": 0.39616289734840393, | |
| "rewards/rejected": -2.422856330871582, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.636887608069164, | |
| "grad_norm": 20.794166929263792, | |
| "learning_rate": 2.199031051284972e-09, | |
| "logits/chosen": -2.007935047149658, | |
| "logits/rejected": -2.0033650398254395, | |
| "logps/chosen": -1.069888710975647, | |
| "logps/rejected": -1.1960642337799072, | |
| "loss": 1.219, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.139777421951294, | |
| "rewards/margins": 0.25235068798065186, | |
| "rewards/rejected": -2.3921284675598145, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 2.6440922190201728, | |
| "grad_norm": 16.98386285768041, | |
| "learning_rate": 2.113872712509254e-09, | |
| "logits/chosen": -1.9919393062591553, | |
| "logits/rejected": -1.9845707416534424, | |
| "logps/chosen": -1.1297125816345215, | |
| "logps/rejected": -1.241287112236023, | |
| "loss": 1.2279, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.259425163269043, | |
| "rewards/margins": 0.22314925491809845, | |
| "rewards/rejected": -2.482574224472046, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 2.6512968299711814, | |
| "grad_norm": 14.064145090241722, | |
| "learning_rate": 2.0303231739801143e-09, | |
| "logits/chosen": -1.9741106033325195, | |
| "logits/rejected": -1.9633283615112305, | |
| "logps/chosen": -1.0185304880142212, | |
| "logps/rejected": -1.1587377786636353, | |
| "loss": 1.1825, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.0370609760284424, | |
| "rewards/margins": 0.28041452169418335, | |
| "rewards/rejected": -2.3174755573272705, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 2.65850144092219, | |
| "grad_norm": 23.56396327392751, | |
| "learning_rate": 1.948388308885102e-09, | |
| "logits/chosen": -2.0366296768188477, | |
| "logits/rejected": -2.02805757522583, | |
| "logps/chosen": -1.0637743473052979, | |
| "logps/rejected": -1.1742548942565918, | |
| "loss": 1.2171, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.1275486946105957, | |
| "rewards/margins": 0.22096149623394012, | |
| "rewards/rejected": -2.3485097885131836, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 2.6657060518731988, | |
| "grad_norm": 25.21273485809688, | |
| "learning_rate": 1.86807387690692e-09, | |
| "logits/chosen": -2.0631988048553467, | |
| "logits/rejected": -2.0600669384002686, | |
| "logps/chosen": -1.0889419317245483, | |
| "logps/rejected": -1.2770618200302124, | |
| "loss": 1.1162, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -2.1778838634490967, | |
| "rewards/margins": 0.3762398660182953, | |
| "rewards/rejected": -2.554123640060425, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.6729106628242074, | |
| "grad_norm": 19.210332180481718, | |
| "learning_rate": 1.789385523818493e-09, | |
| "logits/chosen": -2.027967929840088, | |
| "logits/rejected": -2.0294251441955566, | |
| "logps/chosen": -1.0404349565505981, | |
| "logps/rejected": -1.209099531173706, | |
| "loss": 1.149, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.0808699131011963, | |
| "rewards/margins": 0.33732882142066956, | |
| "rewards/rejected": -2.418199062347412, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 2.680115273775216, | |
| "grad_norm": 25.919412237452388, | |
| "learning_rate": 1.712328781086131e-09, | |
| "logits/chosen": -2.0483665466308594, | |
| "logits/rejected": -2.0430164337158203, | |
| "logps/chosen": -1.1230162382125854, | |
| "logps/rejected": -1.2185190916061401, | |
| "loss": 1.2401, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.246032476425171, | |
| "rewards/margins": 0.191005676984787, | |
| "rewards/rejected": -2.4370381832122803, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 2.6873198847262247, | |
| "grad_norm": 21.180605350865044, | |
| "learning_rate": 1.6369090654806543e-09, | |
| "logits/chosen": -2.0540661811828613, | |
| "logits/rejected": -2.0474164485931396, | |
| "logps/chosen": -1.0206701755523682, | |
| "logps/rejected": -1.1645678281784058, | |
| "loss": 1.1684, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.0413403511047363, | |
| "rewards/margins": 0.2877953350543976, | |
| "rewards/rejected": -2.3291356563568115, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 2.6945244956772334, | |
| "grad_norm": 19.151584962250723, | |
| "learning_rate": 1.5631316786966498e-09, | |
| "logits/chosen": -1.9853427410125732, | |
| "logits/rejected": -1.978816270828247, | |
| "logps/chosen": -1.0220520496368408, | |
| "logps/rejected": -1.1623871326446533, | |
| "loss": 1.1969, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -2.0441040992736816, | |
| "rewards/margins": 0.2806701064109802, | |
| "rewards/rejected": -2.3247742652893066, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 2.701729106628242, | |
| "grad_norm": 18.499060326329523, | |
| "learning_rate": 1.491001806979772e-09, | |
| "logits/chosen": -2.035274028778076, | |
| "logits/rejected": -2.028480052947998, | |
| "logps/chosen": -1.077332854270935, | |
| "logps/rejected": -1.2263195514678955, | |
| "loss": 1.1742, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.15466570854187, | |
| "rewards/margins": 0.29797306656837463, | |
| "rewards/rejected": -2.452639102935791, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.7089337175792507, | |
| "grad_norm": 29.16839407251503, | |
| "learning_rate": 1.4205245207621508e-09, | |
| "logits/chosen": -1.9820353984832764, | |
| "logits/rejected": -1.9796836376190186, | |
| "logps/chosen": -1.1182725429534912, | |
| "logps/rejected": -1.2864872217178345, | |
| "loss": 1.1548, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": -2.2365450859069824, | |
| "rewards/margins": 0.33642950654029846, | |
| "rewards/rejected": -2.572974443435669, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 2.7161383285302594, | |
| "grad_norm": 17.613059928527296, | |
| "learning_rate": 1.3517047743059978e-09, | |
| "logits/chosen": -2.0181725025177, | |
| "logits/rejected": -2.0215516090393066, | |
| "logps/chosen": -1.0734504461288452, | |
| "logps/rejected": -1.2342610359191895, | |
| "loss": 1.1565, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.1469008922576904, | |
| "rewards/margins": 0.3216209411621094, | |
| "rewards/rejected": -2.468522071838379, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 2.723342939481268, | |
| "grad_norm": 17.0753116834011, | |
| "learning_rate": 1.2845474053553156e-09, | |
| "logits/chosen": -2.0134406089782715, | |
| "logits/rejected": -2.0093090534210205, | |
| "logps/chosen": -1.0317740440368652, | |
| "logps/rejected": -1.168919324874878, | |
| "loss": 1.2025, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.0635480880737305, | |
| "rewards/margins": 0.2742905914783478, | |
| "rewards/rejected": -2.337838649749756, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 2.7305475504322767, | |
| "grad_norm": 22.583791154808193, | |
| "learning_rate": 1.2190571347958422e-09, | |
| "logits/chosen": -2.044787883758545, | |
| "logits/rejected": -2.046135187149048, | |
| "logps/chosen": -0.9667074084281921, | |
| "logps/rejected": -1.1686309576034546, | |
| "loss": 1.1099, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -1.9334148168563843, | |
| "rewards/margins": 0.4038470387458801, | |
| "rewards/rejected": -2.337261915206909, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 2.7377521613832854, | |
| "grad_norm": 18.287187828533536, | |
| "learning_rate": 1.1552385663231634e-09, | |
| "logits/chosen": -1.9937756061553955, | |
| "logits/rejected": -1.9841327667236328, | |
| "logps/chosen": -1.0935721397399902, | |
| "logps/rejected": -1.190500020980835, | |
| "loss": 1.2381, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -2.1871442794799805, | |
| "rewards/margins": 0.19385603070259094, | |
| "rewards/rejected": -2.38100004196167, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.744956772334294, | |
| "grad_norm": 18.99456309056716, | |
| "learning_rate": 1.0930961861191302e-09, | |
| "logits/chosen": -1.9580612182617188, | |
| "logits/rejected": -1.9627430438995361, | |
| "logps/chosen": -1.0382286310195923, | |
| "logps/rejected": -1.1804331541061401, | |
| "loss": 1.2008, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -2.0764572620391846, | |
| "rewards/margins": 0.2844088673591614, | |
| "rewards/rejected": -2.3608663082122803, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 2.7521613832853027, | |
| "grad_norm": 16.764883597440075, | |
| "learning_rate": 1.0326343625364608e-09, | |
| "logits/chosen": -1.9691221714019775, | |
| "logits/rejected": -1.9639511108398438, | |
| "logps/chosen": -1.0410795211791992, | |
| "logps/rejected": -1.2136642932891846, | |
| "loss": 1.138, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -2.0821590423583984, | |
| "rewards/margins": 0.3451697826385498, | |
| "rewards/rejected": -2.427328586578369, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 2.7593659942363113, | |
| "grad_norm": 18.44911571731718, | |
| "learning_rate": 9.738573457917066e-10, | |
| "logits/chosen": -2.043980836868286, | |
| "logits/rejected": -2.042267084121704, | |
| "logps/chosen": -1.0499022006988525, | |
| "logps/rejected": -1.2412595748901367, | |
| "loss": 1.11, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": -2.099804401397705, | |
| "rewards/margins": 0.3827148973941803, | |
| "rewards/rejected": -2.4825191497802734, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 2.76657060518732, | |
| "grad_norm": 18.764417824451066, | |
| "learning_rate": 9.16769267666434e-10, | |
| "logits/chosen": -2.0091567039489746, | |
| "logits/rejected": -2.0069632530212402, | |
| "logps/chosen": -1.074094295501709, | |
| "logps/rejected": -1.1498383283615112, | |
| "loss": 1.2637, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -2.148188591003418, | |
| "rewards/margins": 0.15148821473121643, | |
| "rewards/rejected": -2.2996766567230225, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 2.7737752161383287, | |
| "grad_norm": 20.046312375742783, | |
| "learning_rate": 8.613741412168113e-10, | |
| "logits/chosen": -2.027498245239258, | |
| "logits/rejected": -2.026846408843994, | |
| "logps/chosen": -1.0808565616607666, | |
| "logps/rejected": -1.2099745273590088, | |
| "loss": 1.1798, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.161713123321533, | |
| "rewards/margins": 0.2582358717918396, | |
| "rewards/rejected": -2.4199490547180176, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 2.7809798270893373, | |
| "grad_norm": 19.84763607582755, | |
| "learning_rate": 8.076758604914802e-10, | |
| "logits/chosen": -1.957332968711853, | |
| "logits/rejected": -1.9527628421783447, | |
| "logps/chosen": -0.9819733500480652, | |
| "logps/rejected": -1.114538550376892, | |
| "loss": 1.1997, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.9639467000961304, | |
| "rewards/margins": 0.2651303708553314, | |
| "rewards/rejected": -2.229077100753784, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 2.7881844380403455, | |
| "grad_norm": 22.904658084781477, | |
| "learning_rate": 7.55678200257856e-10, | |
| "logits/chosen": -1.9844300746917725, | |
| "logits/rejected": -1.9776723384857178, | |
| "logps/chosen": -1.0327974557876587, | |
| "logps/rejected": -1.1763405799865723, | |
| "loss": 1.1751, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -2.0655949115753174, | |
| "rewards/margins": 0.2870861887931824, | |
| "rewards/rejected": -2.3526811599731445, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 2.795389048991354, | |
| "grad_norm": 17.15291998943784, | |
| "learning_rate": 7.053848157367315e-10, | |
| "logits/chosen": -1.9995343685150146, | |
| "logits/rejected": -1.9940645694732666, | |
| "logps/chosen": -1.0412391424179077, | |
| "logps/rejected": -1.1907306909561157, | |
| "loss": 1.1831, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.0824782848358154, | |
| "rewards/margins": 0.29898306727409363, | |
| "rewards/rejected": -2.3814613819122314, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 2.802593659942363, | |
| "grad_norm": 15.812884819551362, | |
| "learning_rate": 6.567992423453794e-10, | |
| "logits/chosen": -2.0206310749053955, | |
| "logits/rejected": -2.019430637359619, | |
| "logps/chosen": -0.9630235433578491, | |
| "logps/rejected": -1.0794202089309692, | |
| "loss": 1.2021, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -1.9260470867156982, | |
| "rewards/margins": 0.23279304802417755, | |
| "rewards/rejected": -2.1588404178619385, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 2.8097982708933715, | |
| "grad_norm": 19.54993986750196, | |
| "learning_rate": 6.099248954489794e-10, | |
| "logits/chosen": -1.953507423400879, | |
| "logits/rejected": -1.9511181116104126, | |
| "logps/chosen": -1.0681465864181519, | |
| "logps/rejected": -1.229273796081543, | |
| "loss": 1.1568, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.1362931728363037, | |
| "rewards/margins": 0.32225483655929565, | |
| "rewards/rejected": -2.458547592163086, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.81700288184438, | |
| "grad_norm": 22.672929732957467, | |
| "learning_rate": 5.647650701205653e-10, | |
| "logits/chosen": -2.026876449584961, | |
| "logits/rejected": -2.018667697906494, | |
| "logps/chosen": -1.1109135150909424, | |
| "logps/rejected": -1.2674014568328857, | |
| "loss": 1.1766, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.2218270301818848, | |
| "rewards/margins": 0.3129761219024658, | |
| "rewards/rejected": -2.5348029136657715, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 2.824207492795389, | |
| "grad_norm": 16.28695288206369, | |
| "learning_rate": 5.213229409093856e-10, | |
| "logits/chosen": -2.0310721397399902, | |
| "logits/rejected": -2.0254709720611572, | |
| "logps/chosen": -1.05387282371521, | |
| "logps/rejected": -1.1856187582015991, | |
| "loss": 1.2009, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.10774564743042, | |
| "rewards/margins": 0.263491690158844, | |
| "rewards/rejected": -2.3712375164031982, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 2.8314121037463975, | |
| "grad_norm": 20.975683447759703, | |
| "learning_rate": 4.796015616177401e-10, | |
| "logits/chosen": -2.0005106925964355, | |
| "logits/rejected": -1.994783639907837, | |
| "logps/chosen": -1.0665435791015625, | |
| "logps/rejected": -1.1785615682601929, | |
| "loss": 1.2143, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -2.133087158203125, | |
| "rewards/margins": 0.22403590381145477, | |
| "rewards/rejected": -2.3571231365203857, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 2.838616714697406, | |
| "grad_norm": 16.71900486734478, | |
| "learning_rate": 4.3960386508631595e-10, | |
| "logits/chosen": -1.937182068824768, | |
| "logits/rejected": -1.9297128915786743, | |
| "logps/chosen": -0.9666848182678223, | |
| "logps/rejected": -1.0865039825439453, | |
| "loss": 1.2256, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -1.9333696365356445, | |
| "rewards/margins": 0.23963849246501923, | |
| "rewards/rejected": -2.1730079650878906, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 2.845821325648415, | |
| "grad_norm": 35.92105468101964, | |
| "learning_rate": 4.013326629880243e-10, | |
| "logits/chosen": -1.9777085781097412, | |
| "logits/rejected": -1.968076467514038, | |
| "logps/chosen": -1.1062713861465454, | |
| "logps/rejected": -1.2339928150177002, | |
| "loss": 1.2044, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.212542772293091, | |
| "rewards/margins": 0.25544288754463196, | |
| "rewards/rejected": -2.4679856300354004, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 2.8530259365994235, | |
| "grad_norm": 19.697159928360417, | |
| "learning_rate": 3.64790645630339e-10, | |
| "logits/chosen": -1.942647933959961, | |
| "logits/rejected": -1.9421268701553345, | |
| "logps/chosen": -1.0547170639038086, | |
| "logps/rejected": -1.1250708103179932, | |
| "loss": 1.2627, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -2.109434127807617, | |
| "rewards/margins": 0.1407076120376587, | |
| "rewards/rejected": -2.2501416206359863, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 2.860230547550432, | |
| "grad_norm": 21.359517678769173, | |
| "learning_rate": 3.2998038176619e-10, | |
| "logits/chosen": -1.9776138067245483, | |
| "logits/rejected": -1.9692010879516602, | |
| "logps/chosen": -1.0567617416381836, | |
| "logps/rejected": -1.1803498268127441, | |
| "loss": 1.2064, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -2.113523483276367, | |
| "rewards/margins": 0.24717645347118378, | |
| "rewards/rejected": -2.3606996536254883, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 2.867435158501441, | |
| "grad_norm": 20.96219918565088, | |
| "learning_rate": 2.969043184133907e-10, | |
| "logits/chosen": -2.046151638031006, | |
| "logits/rejected": -2.044818639755249, | |
| "logps/chosen": -0.9711786508560181, | |
| "logps/rejected": -1.1876708269119263, | |
| "loss": 1.0771, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -1.9423573017120361, | |
| "rewards/margins": 0.43298429250717163, | |
| "rewards/rejected": -2.3753416538238525, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 2.8746397694524495, | |
| "grad_norm": 17.910920824523004, | |
| "learning_rate": 2.6556478068261447e-10, | |
| "logits/chosen": -1.9732444286346436, | |
| "logits/rejected": -1.9708236455917358, | |
| "logps/chosen": -0.9731259346008301, | |
| "logps/rejected": -1.1022650003433228, | |
| "loss": 1.2084, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -1.9462518692016602, | |
| "rewards/margins": 0.2582783102989197, | |
| "rewards/rejected": -2.2045300006866455, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 2.881844380403458, | |
| "grad_norm": 20.734203195977276, | |
| "learning_rate": 2.3596397161395607e-10, | |
| "logits/chosen": -2.044921875, | |
| "logits/rejected": -2.0331034660339355, | |
| "logps/chosen": -1.0672800540924072, | |
| "logps/rejected": -1.2321101427078247, | |
| "loss": 1.1588, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -2.1345601081848145, | |
| "rewards/margins": 0.32965999841690063, | |
| "rewards/rejected": -2.4642202854156494, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.889048991354467, | |
| "grad_norm": 25.7016655841959, | |
| "learning_rate": 2.0810397202206399e-10, | |
| "logits/chosen": -1.9520553350448608, | |
| "logits/rejected": -1.9573888778686523, | |
| "logps/chosen": -1.063836693763733, | |
| "logps/rejected": -1.193362832069397, | |
| "loss": 1.1905, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.127673387527466, | |
| "rewards/margins": 0.2590521574020386, | |
| "rewards/rejected": -2.386725664138794, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 2.8962536023054755, | |
| "grad_norm": 22.599478343097772, | |
| "learning_rate": 1.819867403498737e-10, | |
| "logits/chosen": -2.0386157035827637, | |
| "logits/rejected": -2.036118984222412, | |
| "logps/chosen": -1.0686347484588623, | |
| "logps/rejected": -1.2007242441177368, | |
| "loss": 1.2018, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.1372694969177246, | |
| "rewards/margins": 0.26417914032936096, | |
| "rewards/rejected": -2.4014484882354736, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 2.903458213256484, | |
| "grad_norm": 21.70269511981427, | |
| "learning_rate": 1.5761411253092382e-10, | |
| "logits/chosen": -1.964998483657837, | |
| "logits/rejected": -1.9548912048339844, | |
| "logps/chosen": -0.9872833490371704, | |
| "logps/rejected": -1.1099205017089844, | |
| "loss": 1.1994, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -1.9745666980743408, | |
| "rewards/margins": 0.24527449905872345, | |
| "rewards/rejected": -2.2198410034179688, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 2.910662824207493, | |
| "grad_norm": 20.259335859045336, | |
| "learning_rate": 1.3498780186031455e-10, | |
| "logits/chosen": -2.010437488555908, | |
| "logits/rejected": -2.0069775581359863, | |
| "logps/chosen": -1.162232756614685, | |
| "logps/rejected": -1.281508207321167, | |
| "loss": 1.2266, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -2.32446551322937, | |
| "rewards/margins": 0.2385510504245758, | |
| "rewards/rejected": -2.563016414642334, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 2.9178674351585014, | |
| "grad_norm": 15.586122569686582, | |
| "learning_rate": 1.1410939887425141e-10, | |
| "logits/chosen": -1.9975839853286743, | |
| "logits/rejected": -1.9993999004364014, | |
| "logps/chosen": -1.045862078666687, | |
| "logps/rejected": -1.1738336086273193, | |
| "loss": 1.212, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -2.091724157333374, | |
| "rewards/margins": 0.25594305992126465, | |
| "rewards/rejected": -2.3476672172546387, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 2.92507204610951, | |
| "grad_norm": 18.548109162992386, | |
| "learning_rate": 9.498037123825686e-11, | |
| "logits/chosen": -2.0100975036621094, | |
| "logits/rejected": -2.0066418647766113, | |
| "logps/chosen": -1.0210684537887573, | |
| "logps/rejected": -1.1468260288238525, | |
| "loss": 1.1987, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -2.0421369075775146, | |
| "rewards/margins": 0.2515150308609009, | |
| "rewards/rejected": -2.293652057647705, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 2.9322766570605188, | |
| "grad_norm": 21.54852206068809, | |
| "learning_rate": 7.760206364398614e-11, | |
| "logits/chosen": -2.0660743713378906, | |
| "logits/rejected": -2.063163995742798, | |
| "logps/chosen": -1.0767936706542969, | |
| "logps/rejected": -1.2189406156539917, | |
| "loss": 1.1849, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.1535873413085938, | |
| "rewards/margins": 0.2842939794063568, | |
| "rewards/rejected": -2.4378812313079834, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 2.9394812680115274, | |
| "grad_norm": 21.178294648611878, | |
| "learning_rate": 6.19756977147029e-11, | |
| "logits/chosen": -1.9951989650726318, | |
| "logits/rejected": -1.9918142557144165, | |
| "logps/chosen": -1.02787446975708, | |
| "logps/rejected": -1.2345163822174072, | |
| "loss": 1.1095, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -2.05574893951416, | |
| "rewards/margins": 0.41328415274620056, | |
| "rewards/rejected": -2.4690327644348145, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 2.946685878962536, | |
| "grad_norm": 20.81149789203122, | |
| "learning_rate": 4.810237191940625e-11, | |
| "logits/chosen": -1.974111557006836, | |
| "logits/rejected": -1.9727462530136108, | |
| "logps/chosen": -1.0376461744308472, | |
| "logps/rejected": -1.1693501472473145, | |
| "loss": 1.217, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -2.0752923488616943, | |
| "rewards/margins": 0.2634081244468689, | |
| "rewards/rejected": -2.338700294494629, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 2.9538904899135447, | |
| "grad_norm": 20.108876799029805, | |
| "learning_rate": 3.5983061495617476e-11, | |
| "logits/chosen": -2.032691240310669, | |
| "logits/rejected": -2.0327444076538086, | |
| "logps/chosen": -1.1233651638031006, | |
| "logps/rejected": -1.2714459896087646, | |
| "loss": 1.1825, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -2.246730327606201, | |
| "rewards/margins": 0.2961619794368744, | |
| "rewards/rejected": -2.5428919792175293, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.9610951008645534, | |
| "grad_norm": 21.51546113795096, | |
| "learning_rate": 2.5618618380812694e-11, | |
| "logits/chosen": -2.0210888385772705, | |
| "logits/rejected": -2.0106148719787598, | |
| "logps/chosen": -1.002300500869751, | |
| "logps/rejected": -1.166154146194458, | |
| "loss": 1.1714, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -2.004601001739502, | |
| "rewards/margins": 0.32770711183547974, | |
| "rewards/rejected": -2.332308292388916, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 2.968299711815562, | |
| "grad_norm": 22.762605833671383, | |
| "learning_rate": 1.700977115254576e-11, | |
| "logits/chosen": -1.9953645467758179, | |
| "logits/rejected": -1.9921376705169678, | |
| "logps/chosen": -0.9968992471694946, | |
| "logps/rejected": -1.1455665826797485, | |
| "loss": 1.1674, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": -1.9937984943389893, | |
| "rewards/margins": 0.29733437299728394, | |
| "rewards/rejected": -2.291133165359497, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 2.9755043227665707, | |
| "grad_norm": 20.444100868277733, | |
| "learning_rate": 1.0157124977230868e-11, | |
| "logits/chosen": -1.9724935293197632, | |
| "logits/rejected": -1.9707790613174438, | |
| "logps/chosen": -0.9694275856018066, | |
| "logps/rejected": -1.117763876914978, | |
| "loss": 1.1687, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -1.9388551712036133, | |
| "rewards/margins": 0.2966724932193756, | |
| "rewards/rejected": -2.235527753829956, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 2.9827089337175794, | |
| "grad_norm": 21.99215491997881, | |
| "learning_rate": 5.061161567596061e-12, | |
| "logits/chosen": -1.9936805963516235, | |
| "logits/rejected": -1.98941171169281, | |
| "logps/chosen": -1.0571701526641846, | |
| "logps/rejected": -1.1420987844467163, | |
| "loss": 1.2615, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -2.114340305328369, | |
| "rewards/margins": 0.16985730826854706, | |
| "rewards/rejected": -2.2841975688934326, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 2.989913544668588, | |
| "grad_norm": 20.878532080632212, | |
| "learning_rate": 1.7222391488297406e-12, | |
| "logits/chosen": -2.013947010040283, | |
| "logits/rejected": -2.010057210922241, | |
| "logps/chosen": -1.1070269346237183, | |
| "logps/rejected": -1.254369854927063, | |
| "loss": 1.1756, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -2.2140538692474365, | |
| "rewards/margins": 0.2946857511997223, | |
| "rewards/rejected": -2.508739709854126, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 2.9971181556195967, | |
| "grad_norm": 19.732213045865922, | |
| "learning_rate": 1.4059243338693238e-13, | |
| "logits/chosen": -1.9882125854492188, | |
| "logits/rejected": -1.9810755252838135, | |
| "logps/chosen": -1.059184193611145, | |
| "logps/rejected": -1.1826164722442627, | |
| "loss": 1.1942, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -2.11836838722229, | |
| "rewards/margins": 0.2468646764755249, | |
| "rewards/rejected": -2.3652329444885254, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 4164, | |
| "total_flos": 0.0, | |
| "train_loss": 1.2025116606473236, | |
| "train_runtime": 6278.9508, | |
| "train_samples_per_second": 10.608, | |
| "train_steps_per_second": 0.663 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 4164, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |