| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2181768596168269, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 9.782608695652175e-07, | |
| "logits/chosen": -1.8868483304977417, | |
| "logits/rejected": -2.3036646842956543, | |
| "logps/chosen": -466.7117004394531, | |
| "logps/rejected": -99.3152084350586, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.956521739130435e-06, | |
| "logits/chosen": -2.2296698093414307, | |
| "logits/rejected": -2.469517469406128, | |
| "logps/chosen": -335.50140380859375, | |
| "logps/rejected": -97.97496032714844, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 2.9347826086956523e-06, | |
| "logits/chosen": -2.125047206878662, | |
| "logits/rejected": -2.445204019546509, | |
| "logps/chosen": -484.7939758300781, | |
| "logps/rejected": -125.93072509765625, | |
| "loss": 1.7268, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -1.0339633226394653, | |
| "rewards/margins": -0.9356753826141357, | |
| "rewards/rejected": -0.09828801453113556, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 3.91304347826087e-06, | |
| "logits/chosen": -1.7763867378234863, | |
| "logits/rejected": -2.1961894035339355, | |
| "logps/chosen": -386.8011779785156, | |
| "logps/rejected": -106.04309844970703, | |
| "loss": 2.3806, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.9238373041152954, | |
| "rewards/margins": -1.2902063131332397, | |
| "rewards/rejected": 0.36636874079704285, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.891304347826087e-06, | |
| "logits/chosen": -1.7257812023162842, | |
| "logits/rejected": -2.353095769882202, | |
| "logps/chosen": -524.7261962890625, | |
| "logps/rejected": -70.26946258544922, | |
| "loss": 0.7488, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": 2.197690010070801, | |
| "rewards/margins": 2.0008764266967773, | |
| "rewards/rejected": 0.19681359827518463, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5.869565217391305e-06, | |
| "logits/chosen": -2.1483359336853027, | |
| "logits/rejected": -2.4817819595336914, | |
| "logps/chosen": -283.6588134765625, | |
| "logps/rejected": -58.25455093383789, | |
| "loss": 0.8712, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": 0.4915351867675781, | |
| "rewards/margins": 0.2005542814731598, | |
| "rewards/rejected": 0.2909809350967407, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 6.847826086956523e-06, | |
| "logits/chosen": -2.1019060611724854, | |
| "logits/rejected": -2.300177812576294, | |
| "logps/chosen": -297.35235595703125, | |
| "logps/rejected": -175.5060272216797, | |
| "loss": 1.1392, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": 0.6018204689025879, | |
| "rewards/margins": -0.036775827407836914, | |
| "rewards/rejected": 0.6385962963104248, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 7.82608695652174e-06, | |
| "logits/chosen": -2.0169122219085693, | |
| "logits/rejected": -2.201702117919922, | |
| "logps/chosen": -342.04852294921875, | |
| "logps/rejected": -117.9761962890625, | |
| "loss": 1.1799, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": 1.111680507659912, | |
| "rewards/margins": 0.17297500371932983, | |
| "rewards/rejected": 0.9387054443359375, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 8.804347826086957e-06, | |
| "logits/chosen": -1.6254596710205078, | |
| "logits/rejected": -2.0724740028381348, | |
| "logps/chosen": -480.1983642578125, | |
| "logps/rejected": -108.05937194824219, | |
| "loss": 0.5203, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": 2.161550283432007, | |
| "rewards/margins": 1.3515225648880005, | |
| "rewards/rejected": 0.8100277185440063, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.782608695652175e-06, | |
| "logits/chosen": -1.4482485055923462, | |
| "logits/rejected": -2.1116294860839844, | |
| "logps/chosen": -451.5759582519531, | |
| "logps/rejected": -67.73944854736328, | |
| "loss": 0.355, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": 6.18800163269043, | |
| "rewards/margins": 5.184471130371094, | |
| "rewards/rejected": 1.0035302639007568, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.0760869565217392e-05, | |
| "logits/chosen": -1.2660062313079834, | |
| "logits/rejected": -1.8649226427078247, | |
| "logps/chosen": -265.8935546875, | |
| "logps/rejected": -57.866336822509766, | |
| "loss": 0.3178, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": 5.78374719619751, | |
| "rewards/margins": 4.856299877166748, | |
| "rewards/rejected": 0.9274475574493408, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.173913043478261e-05, | |
| "logits/chosen": -1.1804625988006592, | |
| "logits/rejected": -1.7009960412979126, | |
| "logps/chosen": -247.2069854736328, | |
| "logps/rejected": -79.96739196777344, | |
| "loss": 0.6617, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": 6.5535197257995605, | |
| "rewards/margins": 4.647799015045166, | |
| "rewards/rejected": 1.9057204723358154, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.2717391304347827e-05, | |
| "logits/chosen": -0.9693112373352051, | |
| "logits/rejected": -1.6497248411178589, | |
| "logps/chosen": -372.8963623046875, | |
| "logps/rejected": -70.47633361816406, | |
| "loss": 0.2498, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": 9.02272891998291, | |
| "rewards/margins": 7.325143814086914, | |
| "rewards/rejected": 1.6975862979888916, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.3695652173913046e-05, | |
| "logits/chosen": -0.7723280787467957, | |
| "logits/rejected": -1.4176554679870605, | |
| "logps/chosen": -318.0392150878906, | |
| "logps/rejected": -79.94412994384766, | |
| "loss": 0.6401, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": 6.698611259460449, | |
| "rewards/margins": 4.517416477203369, | |
| "rewards/rejected": 2.18119478225708, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.4673913043478261e-05, | |
| "logits/chosen": -0.8084051609039307, | |
| "logits/rejected": -1.4428132772445679, | |
| "logps/chosen": -363.15869140625, | |
| "logps/rejected": -70.70079803466797, | |
| "loss": 0.4963, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": 7.02784538269043, | |
| "rewards/margins": 5.400259971618652, | |
| "rewards/rejected": 1.6275854110717773, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.565217391304348e-05, | |
| "logits/chosen": -0.8056433200836182, | |
| "logits/rejected": -1.4629294872283936, | |
| "logps/chosen": -328.90435791015625, | |
| "logps/rejected": -77.36820983886719, | |
| "loss": 0.544, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": 9.417024612426758, | |
| "rewards/margins": 7.023505687713623, | |
| "rewards/rejected": 2.393519163131714, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.6630434782608694e-05, | |
| "logits/chosen": -1.0918309688568115, | |
| "logits/rejected": -1.6613047122955322, | |
| "logps/chosen": -277.37982177734375, | |
| "logps/rejected": -56.75635528564453, | |
| "loss": 0.8958, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": 6.844330310821533, | |
| "rewards/margins": 4.82295560836792, | |
| "rewards/rejected": 2.0213749408721924, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.7608695652173915e-05, | |
| "logits/chosen": -0.9810999631881714, | |
| "logits/rejected": -1.975247859954834, | |
| "logps/chosen": -465.6856689453125, | |
| "logps/rejected": -56.440269470214844, | |
| "loss": 0.2275, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": 7.767542362213135, | |
| "rewards/margins": 6.572459697723389, | |
| "rewards/rejected": 1.1950825452804565, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.8586956521739132e-05, | |
| "logits/chosen": -1.2428462505340576, | |
| "logits/rejected": -1.8064367771148682, | |
| "logps/chosen": -322.7100830078125, | |
| "logps/rejected": -108.71321105957031, | |
| "loss": 0.3006, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": 6.459216594696045, | |
| "rewards/margins": 3.9207711219787598, | |
| "rewards/rejected": 2.5384464263916016, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.956521739130435e-05, | |
| "logits/chosen": -1.7773098945617676, | |
| "logits/rejected": -2.178387403488159, | |
| "logps/chosen": -357.8656921386719, | |
| "logps/rejected": -112.65576934814453, | |
| "loss": 0.3837, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 4.684590816497803, | |
| "rewards/margins": 4.16137170791626, | |
| "rewards/rejected": 0.5232191681861877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.0543478260869567e-05, | |
| "logits/chosen": -1.6802027225494385, | |
| "logits/rejected": -1.943274974822998, | |
| "logps/chosen": -327.85693359375, | |
| "logps/rejected": -87.53279113769531, | |
| "loss": 0.4428, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 2.681103229522705, | |
| "rewards/margins": 2.3518502712249756, | |
| "rewards/rejected": 0.32925307750701904, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.1521739130434784e-05, | |
| "logits/chosen": -1.5789223909378052, | |
| "logits/rejected": -2.4593281745910645, | |
| "logps/chosen": -590.5732421875, | |
| "logps/rejected": -86.78819274902344, | |
| "loss": 0.252, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": 3.6503655910491943, | |
| "rewards/margins": 3.621542453765869, | |
| "rewards/rejected": 0.028822531923651695, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.25e-05, | |
| "logits/chosen": -1.8142305612564087, | |
| "logits/rejected": -2.3507237434387207, | |
| "logps/chosen": -296.42266845703125, | |
| "logps/rejected": -60.37131118774414, | |
| "loss": 0.5602, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": 2.170171022415161, | |
| "rewards/margins": 2.5353941917419434, | |
| "rewards/rejected": -0.36522313952445984, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.347826086956522e-05, | |
| "logits/chosen": -1.7050156593322754, | |
| "logits/rejected": -2.158708333969116, | |
| "logps/chosen": -523.41748046875, | |
| "logps/rejected": -81.6954345703125, | |
| "loss": 0.7359, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 2.3799068927764893, | |
| "rewards/margins": 2.963341236114502, | |
| "rewards/rejected": -0.583433985710144, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.4456521739130436e-05, | |
| "logits/chosen": -1.704190731048584, | |
| "logits/rejected": -2.215942859649658, | |
| "logps/chosen": -457.6519775390625, | |
| "logps/rejected": -91.778076171875, | |
| "loss": 0.0865, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.57167911529541, | |
| "rewards/margins": 4.6597795486450195, | |
| "rewards/rejected": -0.08810039609670639, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.5434782608695653e-05, | |
| "logits/chosen": -1.4195502996444702, | |
| "logits/rejected": -2.109910488128662, | |
| "logps/chosen": -445.05908203125, | |
| "logps/rejected": -80.95398712158203, | |
| "loss": 0.1265, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 5.362960338592529, | |
| "rewards/margins": 5.412578582763672, | |
| "rewards/rejected": -0.049618080258369446, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.6413043478260874e-05, | |
| "logits/chosen": -1.0274971723556519, | |
| "logits/rejected": -1.7681918144226074, | |
| "logps/chosen": -393.85870361328125, | |
| "logps/rejected": -77.79011535644531, | |
| "loss": 0.1702, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.572484016418457, | |
| "rewards/margins": 4.946743488311768, | |
| "rewards/rejected": -0.374259889125824, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.739130434782609e-05, | |
| "logits/chosen": -0.7365273237228394, | |
| "logits/rejected": -1.9988534450531006, | |
| "logps/chosen": -569.8936767578125, | |
| "logps/rejected": -79.71562194824219, | |
| "loss": 0.0465, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 12.751686096191406, | |
| "rewards/margins": 12.474647521972656, | |
| "rewards/rejected": 0.2770393192768097, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.836956521739131e-05, | |
| "logits/chosen": -0.6424872279167175, | |
| "logits/rejected": -1.6069284677505493, | |
| "logps/chosen": -368.68292236328125, | |
| "logps/rejected": -103.9141845703125, | |
| "loss": 0.0351, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.144453048706055, | |
| "rewards/margins": 10.381340980529785, | |
| "rewards/rejected": -0.23688830435276031, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.9347826086956523e-05, | |
| "logits/chosen": -0.6061868071556091, | |
| "logits/rejected": -1.4256294965744019, | |
| "logps/chosen": -360.4029846191406, | |
| "logps/rejected": -130.27224731445312, | |
| "loss": 0.0414, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 12.649539947509766, | |
| "rewards/margins": 12.36604118347168, | |
| "rewards/rejected": 0.2834990620613098, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.032608695652174e-05, | |
| "logits/chosen": -1.0075111389160156, | |
| "logits/rejected": -1.3918204307556152, | |
| "logps/chosen": -227.05450439453125, | |
| "logps/rejected": -172.67947387695312, | |
| "loss": 1.1145, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": 4.628115653991699, | |
| "rewards/margins": 3.9521055221557617, | |
| "rewards/rejected": 0.6760098934173584, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.130434782608696e-05, | |
| "logits/chosen": -0.9060839414596558, | |
| "logits/rejected": -2.0123891830444336, | |
| "logps/chosen": -403.06781005859375, | |
| "logps/rejected": -75.35684967041016, | |
| "loss": 0.0319, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.352475643157959, | |
| "rewards/margins": 8.778717041015625, | |
| "rewards/rejected": -1.4262410402297974, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.228260869565217e-05, | |
| "logits/chosen": -1.2984271049499512, | |
| "logits/rejected": -1.9507720470428467, | |
| "logps/chosen": -306.7958984375, | |
| "logps/rejected": -107.01404571533203, | |
| "loss": 0.0795, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 5.223363399505615, | |
| "rewards/margins": 7.012631416320801, | |
| "rewards/rejected": -1.7892680168151855, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.326086956521739e-05, | |
| "logits/chosen": -1.1749062538146973, | |
| "logits/rejected": -2.004805564880371, | |
| "logps/chosen": -437.9205322265625, | |
| "logps/rejected": -135.21376037597656, | |
| "loss": 0.0021, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.3555908203125, | |
| "rewards/margins": 12.285521507263184, | |
| "rewards/rejected": -2.9299299716949463, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.423913043478261e-05, | |
| "logits/chosen": -0.9383536577224731, | |
| "logits/rejected": -2.0366413593292236, | |
| "logps/chosen": -457.8018493652344, | |
| "logps/rejected": -87.7723617553711, | |
| "loss": 0.041, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.836911678314209, | |
| "rewards/margins": 9.43667984008789, | |
| "rewards/rejected": -2.599769115447998, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.521739130434783e-05, | |
| "logits/chosen": -1.120687484741211, | |
| "logits/rejected": -1.9933563470840454, | |
| "logps/chosen": -439.6186828613281, | |
| "logps/rejected": -94.24763488769531, | |
| "loss": 0.1906, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 6.565365314483643, | |
| "rewards/margins": 9.163484573364258, | |
| "rewards/rejected": -2.5981192588806152, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.619565217391305e-05, | |
| "logits/chosen": -1.338189959526062, | |
| "logits/rejected": -1.9118335247039795, | |
| "logps/chosen": -286.96728515625, | |
| "logps/rejected": -127.22909545898438, | |
| "loss": 0.0279, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9341468811035156, | |
| "rewards/margins": 6.362301349639893, | |
| "rewards/rejected": -3.4281537532806396, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.7173913043478264e-05, | |
| "logits/chosen": -1.2371840476989746, | |
| "logits/rejected": -1.7362310886383057, | |
| "logps/chosen": -319.52911376953125, | |
| "logps/rejected": -151.65084838867188, | |
| "loss": 0.0939, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 6.216956615447998, | |
| "rewards/margins": 9.073128700256348, | |
| "rewards/rejected": -2.856172561645508, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.815217391304348e-05, | |
| "logits/chosen": -0.8356418609619141, | |
| "logits/rejected": -1.7252635955810547, | |
| "logps/chosen": -391.63275146484375, | |
| "logps/rejected": -154.64361572265625, | |
| "loss": 0.2483, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 8.802470207214355, | |
| "rewards/margins": 11.676048278808594, | |
| "rewards/rejected": -2.873577356338501, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.91304347826087e-05, | |
| "logits/chosen": -0.9311200380325317, | |
| "logits/rejected": -1.8107151985168457, | |
| "logps/chosen": -374.2525939941406, | |
| "logps/rejected": -131.86715698242188, | |
| "loss": 0.0058, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.586356163024902, | |
| "rewards/margins": 11.15583324432373, | |
| "rewards/rejected": -2.569479465484619, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.0108695652173916e-05, | |
| "logits/chosen": -0.6861732006072998, | |
| "logits/rejected": -1.6295411586761475, | |
| "logps/chosen": -311.7944641113281, | |
| "logps/rejected": -132.4564666748047, | |
| "loss": 0.0124, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.333560943603516, | |
| "rewards/margins": 10.023443222045898, | |
| "rewards/rejected": -4.689882755279541, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.1086956521739134e-05, | |
| "logits/chosen": -1.1133558750152588, | |
| "logits/rejected": -2.0819039344787598, | |
| "logps/chosen": -282.22332763671875, | |
| "logps/rejected": -98.7315673828125, | |
| "loss": 0.0156, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.930261611938477, | |
| "rewards/margins": 8.344033241271973, | |
| "rewards/rejected": -3.413771629333496, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.206521739130435e-05, | |
| "logits/chosen": -1.0311042070388794, | |
| "logits/rejected": -2.0758776664733887, | |
| "logps/chosen": -369.7279052734375, | |
| "logps/rejected": -146.08364868164062, | |
| "loss": 0.0127, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.7552289962768555, | |
| "rewards/margins": 13.249262809753418, | |
| "rewards/rejected": -5.494032859802246, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.304347826086957e-05, | |
| "logits/chosen": -1.0191632509231567, | |
| "logits/rejected": -1.995982050895691, | |
| "logps/chosen": -367.692626953125, | |
| "logps/rejected": -152.65135192871094, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.9517822265625, | |
| "rewards/margins": 13.869260787963867, | |
| "rewards/rejected": -5.917478084564209, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.4021739130434786e-05, | |
| "logits/chosen": -1.3430471420288086, | |
| "logits/rejected": -2.0910849571228027, | |
| "logps/chosen": -334.8968811035156, | |
| "logps/rejected": -168.4519500732422, | |
| "loss": 0.0057, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.113959312438965, | |
| "rewards/margins": 12.524876594543457, | |
| "rewards/rejected": -7.41091775894165, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.5e-05, | |
| "logits/chosen": -0.8516018986701965, | |
| "logits/rejected": -1.9934579133987427, | |
| "logps/chosen": -386.7164001464844, | |
| "logps/rejected": -152.8460235595703, | |
| "loss": 0.0253, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.0803022384643555, | |
| "rewards/margins": 13.202106475830078, | |
| "rewards/rejected": -6.121803283691406, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.494827586206897e-05, | |
| "logits/chosen": -0.9460695385932922, | |
| "logits/rejected": -1.9993284940719604, | |
| "logps/chosen": -355.2873229980469, | |
| "logps/rejected": -188.5989532470703, | |
| "loss": 0.1008, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 6.138028144836426, | |
| "rewards/margins": 13.93982982635498, | |
| "rewards/rejected": -7.801800727844238, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.489655172413793e-05, | |
| "logits/chosen": -1.212510108947754, | |
| "logits/rejected": -2.0202791690826416, | |
| "logps/chosen": -293.0318908691406, | |
| "logps/rejected": -194.92649841308594, | |
| "loss": 0.0058, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.742659091949463, | |
| "rewards/margins": 17.209505081176758, | |
| "rewards/rejected": -10.466848373413086, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.48448275862069e-05, | |
| "logits/chosen": -1.388157606124878, | |
| "logits/rejected": -2.1885251998901367, | |
| "logps/chosen": -330.6868591308594, | |
| "logps/rejected": -214.35423278808594, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.653369903564453, | |
| "rewards/margins": 18.257244110107422, | |
| "rewards/rejected": -12.603873252868652, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.479310344827587e-05, | |
| "logits/chosen": -1.020527958869934, | |
| "logits/rejected": -1.9433815479278564, | |
| "logps/chosen": -347.48394775390625, | |
| "logps/rejected": -191.15147399902344, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.30579137802124, | |
| "rewards/margins": 18.734811782836914, | |
| "rewards/rejected": -11.429021835327148, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.474137931034483e-05, | |
| "logits/chosen": -1.2260757684707642, | |
| "logits/rejected": -2.0841352939605713, | |
| "logps/chosen": -243.5459747314453, | |
| "logps/rejected": -188.00442504882812, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.675260543823242, | |
| "rewards/margins": 14.774760246276855, | |
| "rewards/rejected": -11.099499702453613, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.46896551724138e-05, | |
| "logits/chosen": -1.0378711223602295, | |
| "logits/rejected": -2.0174639225006104, | |
| "logps/chosen": -375.9520263671875, | |
| "logps/rejected": -182.97463989257812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.18237829208374, | |
| "rewards/margins": 19.010208129882812, | |
| "rewards/rejected": -11.82783317565918, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.4637931034482765e-05, | |
| "logits/chosen": -1.1728301048278809, | |
| "logits/rejected": -1.9667140245437622, | |
| "logps/chosen": -309.5382995605469, | |
| "logps/rejected": -241.13743591308594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.681103706359863, | |
| "rewards/margins": 21.542236328125, | |
| "rewards/rejected": -14.861133575439453, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.4586206896551726e-05, | |
| "logits/chosen": -1.1491725444793701, | |
| "logits/rejected": -1.760243535041809, | |
| "logps/chosen": -250.09036254882812, | |
| "logps/rejected": -226.221435546875, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.3173880577087402, | |
| "rewards/margins": 16.417604446411133, | |
| "rewards/rejected": -14.100215911865234, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.4534482758620694e-05, | |
| "logits/chosen": -1.2637637853622437, | |
| "logits/rejected": -1.8996949195861816, | |
| "logps/chosen": -401.9977722167969, | |
| "logps/rejected": -303.9198303222656, | |
| "loss": 0.0011, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.595758438110352, | |
| "rewards/margins": 26.598325729370117, | |
| "rewards/rejected": -18.002567291259766, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.4482758620689656e-05, | |
| "logits/chosen": -0.9855178594589233, | |
| "logits/rejected": -2.0190346240997314, | |
| "logps/chosen": -394.40625, | |
| "logps/rejected": -253.97720336914062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.395060062408447, | |
| "rewards/margins": 23.680639266967773, | |
| "rewards/rejected": -17.28557777404785, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.4431034482758624e-05, | |
| "logits/chosen": -1.049959421157837, | |
| "logits/rejected": -1.6531357765197754, | |
| "logps/chosen": -205.54058837890625, | |
| "logps/rejected": -250.66958618164062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.874389171600342, | |
| "rewards/margins": 20.083953857421875, | |
| "rewards/rejected": -15.209564208984375, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.4379310344827585e-05, | |
| "logits/chosen": -0.8982111811637878, | |
| "logits/rejected": -1.6001371145248413, | |
| "logps/chosen": -241.91790771484375, | |
| "logps/rejected": -252.10299682617188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.044814109802246, | |
| "rewards/margins": 21.77715301513672, | |
| "rewards/rejected": -15.732340812683105, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.432758620689655e-05, | |
| "logits/chosen": -0.753044068813324, | |
| "logits/rejected": -1.6325318813323975, | |
| "logps/chosen": -315.30230712890625, | |
| "logps/rejected": -245.91976928710938, | |
| "loss": 0.0078, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.801758289337158, | |
| "rewards/margins": 20.689924240112305, | |
| "rewards/rejected": -14.888164520263672, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.427586206896552e-05, | |
| "logits/chosen": -0.5443448424339294, | |
| "logits/rejected": -1.7224345207214355, | |
| "logps/chosen": -388.339599609375, | |
| "logps/rejected": -211.5500946044922, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 11.5880708694458, | |
| "rewards/margins": 24.889877319335938, | |
| "rewards/rejected": -13.301810264587402, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.422413793103448e-05, | |
| "logits/chosen": -0.6921290755271912, | |
| "logits/rejected": -1.6139135360717773, | |
| "logps/chosen": -242.47312927246094, | |
| "logps/rejected": -214.73622131347656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 11.329044342041016, | |
| "rewards/margins": 21.45581817626953, | |
| "rewards/rejected": -10.1267728805542, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.417241379310345e-05, | |
| "logits/chosen": -0.4828271269798279, | |
| "logits/rejected": -1.649349570274353, | |
| "logps/chosen": -306.6004638671875, | |
| "logps/rejected": -172.2523956298828, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 11.373737335205078, | |
| "rewards/margins": 20.521434783935547, | |
| "rewards/rejected": -9.147698402404785, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.412068965517242e-05, | |
| "logits/chosen": -0.4676312506198883, | |
| "logits/rejected": -1.6393827199935913, | |
| "logps/chosen": -357.127685546875, | |
| "logps/rejected": -199.84524536132812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 16.730457305908203, | |
| "rewards/margins": 26.409488677978516, | |
| "rewards/rejected": -9.679031372070312, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.406896551724138e-05, | |
| "logits/chosen": -0.7774850726127625, | |
| "logits/rejected": -1.324657678604126, | |
| "logps/chosen": -193.75833129882812, | |
| "logps/rejected": -194.70233154296875, | |
| "loss": 0.0015, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.371121406555176, | |
| "rewards/margins": 15.129293441772461, | |
| "rewards/rejected": -9.758172988891602, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.401724137931035e-05, | |
| "logits/chosen": -0.5502160787582397, | |
| "logits/rejected": -1.466262936592102, | |
| "logps/chosen": -240.96836853027344, | |
| "logps/rejected": -174.6772003173828, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.808657169342041, | |
| "rewards/margins": 17.520769119262695, | |
| "rewards/rejected": -9.712109565734863, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.3965517241379315e-05, | |
| "logits/chosen": -0.5615445375442505, | |
| "logits/rejected": -1.3243337869644165, | |
| "logps/chosen": -282.65484619140625, | |
| "logps/rejected": -211.33326721191406, | |
| "loss": 0.4403, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 9.213937759399414, | |
| "rewards/margins": 18.93706512451172, | |
| "rewards/rejected": -9.723125457763672, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.3913793103448277e-05, | |
| "logits/chosen": -0.5882927775382996, | |
| "logits/rejected": -1.4038889408111572, | |
| "logps/chosen": -244.26333618164062, | |
| "logps/rejected": -192.05517578125, | |
| "loss": 0.0007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.999594688415527, | |
| "rewards/margins": 16.232702255249023, | |
| "rewards/rejected": -10.233107566833496, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.3862068965517245e-05, | |
| "logits/chosen": -0.7710579633712769, | |
| "logits/rejected": -1.5811477899551392, | |
| "logps/chosen": -274.4286804199219, | |
| "logps/rejected": -189.31427001953125, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.483807563781738, | |
| "rewards/margins": 22.807409286499023, | |
| "rewards/rejected": -12.323600769042969, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.381034482758621e-05, | |
| "logits/chosen": -0.2001597136259079, | |
| "logits/rejected": -1.5477687120437622, | |
| "logps/chosen": -398.04296875, | |
| "logps/rejected": -208.8218536376953, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 11.122913360595703, | |
| "rewards/margins": 23.094343185424805, | |
| "rewards/rejected": -11.971429824829102, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.3758620689655174e-05, | |
| "logits/chosen": -0.5891858339309692, | |
| "logits/rejected": -1.500455379486084, | |
| "logps/chosen": -223.07235717773438, | |
| "logps/rejected": -219.37405395507812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.498288154602051, | |
| "rewards/margins": 19.65944480895996, | |
| "rewards/rejected": -12.16115665435791, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.370689655172414e-05, | |
| "logits/chosen": -0.5222135186195374, | |
| "logits/rejected": -1.459835410118103, | |
| "logps/chosen": -264.3045349121094, | |
| "logps/rejected": -184.4599609375, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.261447429656982, | |
| "rewards/margins": 17.37218475341797, | |
| "rewards/rejected": -10.110737800598145, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.365517241379311e-05, | |
| "logits/chosen": -0.4974746108055115, | |
| "logits/rejected": -1.5458664894104004, | |
| "logps/chosen": -272.7602233886719, | |
| "logps/rejected": -181.00247192382812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.925564765930176, | |
| "rewards/margins": 22.81020736694336, | |
| "rewards/rejected": -11.88464069366455, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.360344827586207e-05, | |
| "logits/chosen": -0.3357059359550476, | |
| "logits/rejected": -1.4441150426864624, | |
| "logps/chosen": -414.8647766113281, | |
| "logps/rejected": -223.109375, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 13.181656837463379, | |
| "rewards/margins": 27.527368545532227, | |
| "rewards/rejected": -14.345712661743164, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.355172413793104e-05, | |
| "logits/chosen": -0.6531690359115601, | |
| "logits/rejected": -1.4611525535583496, | |
| "logps/chosen": -221.1251220703125, | |
| "logps/rejected": -226.30352783203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.807477951049805, | |
| "rewards/margins": 21.502174377441406, | |
| "rewards/rejected": -13.694696426391602, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.35e-05, | |
| "logits/chosen": -0.5135932564735413, | |
| "logits/rejected": -1.4995931386947632, | |
| "logps/chosen": -332.56842041015625, | |
| "logps/rejected": -270.9891662597656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 12.301117897033691, | |
| "rewards/margins": 26.830467224121094, | |
| "rewards/rejected": -14.529346466064453, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.344827586206897e-05, | |
| "logits/chosen": -0.3770020604133606, | |
| "logits/rejected": -1.44219970703125, | |
| "logps/chosen": -317.96240234375, | |
| "logps/rejected": -250.3682098388672, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.8860931396484375, | |
| "rewards/margins": 23.53211784362793, | |
| "rewards/rejected": -15.646024703979492, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.339655172413793e-05, | |
| "logits/chosen": -0.49628371000289917, | |
| "logits/rejected": -1.4045476913452148, | |
| "logps/chosen": -367.6162414550781, | |
| "logps/rejected": -265.2467346191406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.148900985717773, | |
| "rewards/margins": 24.313669204711914, | |
| "rewards/rejected": -14.164766311645508, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.33448275862069e-05, | |
| "logits/chosen": -0.5287263989448547, | |
| "logits/rejected": -1.3665097951889038, | |
| "logps/chosen": -341.87664794921875, | |
| "logps/rejected": -249.6795654296875, | |
| "loss": 0.1517, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 8.620326042175293, | |
| "rewards/margins": 19.850658416748047, | |
| "rewards/rejected": -11.230331420898438, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.3293103448275865e-05, | |
| "logits/chosen": -0.2743966281414032, | |
| "logits/rejected": -1.4255130290985107, | |
| "logps/chosen": -334.3340759277344, | |
| "logps/rejected": -265.15032958984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.872364044189453, | |
| "rewards/margins": 27.00977325439453, | |
| "rewards/rejected": -16.13741111755371, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.324137931034483e-05, | |
| "logits/chosen": -0.5563563108444214, | |
| "logits/rejected": -1.6934064626693726, | |
| "logps/chosen": -408.9612731933594, | |
| "logps/rejected": -220.00091552734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.237764358520508, | |
| "rewards/margins": 24.824390411376953, | |
| "rewards/rejected": -14.586627960205078, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.3189655172413795e-05, | |
| "logits/chosen": -0.617451548576355, | |
| "logits/rejected": -1.640676736831665, | |
| "logps/chosen": -363.1540222167969, | |
| "logps/rejected": -226.42376708984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.672024726867676, | |
| "rewards/margins": 24.347591400146484, | |
| "rewards/rejected": -13.675565719604492, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.313793103448276e-05, | |
| "logits/chosen": -0.45464372634887695, | |
| "logits/rejected": -1.4708175659179688, | |
| "logps/chosen": -328.35675048828125, | |
| "logps/rejected": -223.79766845703125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.965122699737549, | |
| "rewards/margins": 21.22077178955078, | |
| "rewards/rejected": -14.25564956665039, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.3086206896551724e-05, | |
| "logits/chosen": -0.6079340577125549, | |
| "logits/rejected": -1.5333082675933838, | |
| "logps/chosen": -219.4099578857422, | |
| "logps/rejected": -230.07948303222656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.278050899505615, | |
| "rewards/margins": 20.45656967163086, | |
| "rewards/rejected": -14.178520202636719, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.303448275862069e-05, | |
| "logits/chosen": -0.4874611794948578, | |
| "logits/rejected": -1.5135740041732788, | |
| "logps/chosen": -304.00274658203125, | |
| "logps/rejected": -263.0120849609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.557435989379883, | |
| "rewards/margins": 24.41012954711914, | |
| "rewards/rejected": -14.852693557739258, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.298275862068966e-05, | |
| "logits/chosen": -0.4802761375904083, | |
| "logits/rejected": -1.4294252395629883, | |
| "logps/chosen": -293.41192626953125, | |
| "logps/rejected": -283.2821350097656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.370968818664551, | |
| "rewards/margins": 24.32880401611328, | |
| "rewards/rejected": -16.957834243774414, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.293103448275862e-05, | |
| "logits/chosen": -0.3580012321472168, | |
| "logits/rejected": -1.499201774597168, | |
| "logps/chosen": -350.5351257324219, | |
| "logps/rejected": -244.6523895263672, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 12.237674713134766, | |
| "rewards/margins": 26.58084487915039, | |
| "rewards/rejected": -14.343170166015625, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.287931034482759e-05, | |
| "logits/chosen": -0.6917211413383484, | |
| "logits/rejected": -1.4503819942474365, | |
| "logps/chosen": -217.5175323486328, | |
| "logps/rejected": -228.9481201171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.635045051574707, | |
| "rewards/margins": 20.42833709716797, | |
| "rewards/rejected": -13.793292999267578, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.282758620689656e-05, | |
| "logits/chosen": -0.8255922794342041, | |
| "logits/rejected": -1.4748644828796387, | |
| "logps/chosen": -217.336669921875, | |
| "logps/rejected": -206.8959197998047, | |
| "loss": 0.0005, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.22888708114624, | |
| "rewards/margins": 19.673564910888672, | |
| "rewards/rejected": -14.444679260253906, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2775862068965525e-05, | |
| "logits/chosen": -0.626908004283905, | |
| "logits/rejected": -1.6663663387298584, | |
| "logps/chosen": -328.56231689453125, | |
| "logps/rejected": -235.42843627929688, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.891221046447754, | |
| "rewards/margins": 21.77893829345703, | |
| "rewards/rejected": -14.887716293334961, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2724137931034486e-05, | |
| "logits/chosen": -0.5689429640769958, | |
| "logits/rejected": -1.618609070777893, | |
| "logps/chosen": -298.08770751953125, | |
| "logps/rejected": -237.06094360351562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.328791618347168, | |
| "rewards/margins": 24.572046279907227, | |
| "rewards/rejected": -15.243253707885742, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2672413793103454e-05, | |
| "logits/chosen": -0.5474239587783813, | |
| "logits/rejected": -1.6142076253890991, | |
| "logps/chosen": -295.00494384765625, | |
| "logps/rejected": -268.9505615234375, | |
| "loss": 0.0287, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.715471267700195, | |
| "rewards/margins": 23.968780517578125, | |
| "rewards/rejected": -14.25330924987793, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2620689655172416e-05, | |
| "logits/chosen": -0.5132906436920166, | |
| "logits/rejected": -1.6248977184295654, | |
| "logps/chosen": -314.5284423828125, | |
| "logps/rejected": -210.1342315673828, | |
| "loss": 0.0595, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 7.707674980163574, | |
| "rewards/margins": 20.18167495727539, | |
| "rewards/rejected": -12.474000930786133, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2568965517241384e-05, | |
| "logits/chosen": -0.45244526863098145, | |
| "logits/rejected": -1.7539148330688477, | |
| "logps/chosen": -394.9619140625, | |
| "logps/rejected": -230.32550048828125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 11.32716178894043, | |
| "rewards/margins": 26.326129913330078, | |
| "rewards/rejected": -14.998967170715332, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2517241379310345e-05, | |
| "logits/chosen": -0.7538139224052429, | |
| "logits/rejected": -1.5796599388122559, | |
| "logps/chosen": -286.95281982421875, | |
| "logps/rejected": -266.7412414550781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.003694534301758, | |
| "rewards/margins": 25.05299949645996, | |
| "rewards/rejected": -17.04930305480957, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.246551724137931e-05, | |
| "logits/chosen": -0.44376400113105774, | |
| "logits/rejected": -1.5493687391281128, | |
| "logps/chosen": -406.93951416015625, | |
| "logps/rejected": -302.9401550292969, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.246862888336182, | |
| "rewards/margins": 27.937149047851562, | |
| "rewards/rejected": -20.69028663635254, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.2413793103448274e-05, | |
| "logits/chosen": -0.7577255964279175, | |
| "logits/rejected": -1.77558171749115, | |
| "logps/chosen": -357.90838623046875, | |
| "logps/rejected": -246.73458862304688, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.267894744873047, | |
| "rewards/margins": 24.79804229736328, | |
| "rewards/rejected": -17.530147552490234, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.236206896551724e-05, | |
| "logits/chosen": -0.8565713763237, | |
| "logits/rejected": -1.7760589122772217, | |
| "logps/chosen": -338.751220703125, | |
| "logps/rejected": -304.101806640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.238365173339844, | |
| "rewards/margins": 27.781978607177734, | |
| "rewards/rejected": -18.54361343383789, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.231034482758621e-05, | |
| "logits/chosen": -0.40401750802993774, | |
| "logits/rejected": -1.7701493501663208, | |
| "logps/chosen": -537.5440063476562, | |
| "logps/rejected": -349.8548889160156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 14.211512565612793, | |
| "rewards/margins": 37.746315002441406, | |
| "rewards/rejected": -23.53480339050293, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.225862068965517e-05, | |
| "logits/chosen": -0.9799883365631104, | |
| "logits/rejected": -1.8936114311218262, | |
| "logps/chosen": -403.7567138671875, | |
| "logps/rejected": -257.6401672363281, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.710639476776123, | |
| "rewards/margins": 23.923961639404297, | |
| "rewards/rejected": -19.213321685791016, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.220689655172414e-05, | |
| "logits/chosen": -0.94623202085495, | |
| "logits/rejected": -1.7190004587173462, | |
| "logps/chosen": -360.8213806152344, | |
| "logps/rejected": -310.1994323730469, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.991710662841797, | |
| "rewards/margins": 26.59444808959961, | |
| "rewards/rejected": -22.60273551940918, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.215517241379311e-05, | |
| "logits/chosen": -0.6102348566055298, | |
| "logits/rejected": -1.8472094535827637, | |
| "logps/chosen": -460.48699951171875, | |
| "logps/rejected": -369.1959228515625, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.696294784545898, | |
| "rewards/margins": 32.79343795776367, | |
| "rewards/rejected": -26.097143173217773, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.210344827586207e-05, | |
| "logits/chosen": -0.9649127721786499, | |
| "logits/rejected": -1.8413913249969482, | |
| "logps/chosen": -455.1119079589844, | |
| "logps/rejected": -365.0443420410156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.14402437210083, | |
| "rewards/margins": 32.23406219482422, | |
| "rewards/rejected": -25.090038299560547, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.2051724137931036e-05, | |
| "logits/chosen": -1.176360011100769, | |
| "logits/rejected": -1.6914632320404053, | |
| "logps/chosen": -283.3887023925781, | |
| "logps/rejected": -343.1808166503906, | |
| "loss": 0.0008, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.03441965579986572, | |
| "rewards/margins": 24.389293670654297, | |
| "rewards/rejected": -24.423709869384766, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.2000000000000004e-05, | |
| "logits/chosen": -0.7725791335105896, | |
| "logits/rejected": -1.943673014640808, | |
| "logps/chosen": -484.6883544921875, | |
| "logps/rejected": -338.5666198730469, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.784733295440674, | |
| "rewards/margins": 31.62430763244629, | |
| "rewards/rejected": -23.839576721191406, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.194827586206897e-05, | |
| "logits/chosen": -0.6000739336013794, | |
| "logits/rejected": -1.681613802909851, | |
| "logps/chosen": -342.3570556640625, | |
| "logps/rejected": -256.06829833984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.42122745513916, | |
| "rewards/margins": 25.636981964111328, | |
| "rewards/rejected": -18.21575355529785, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.1896551724137934e-05, | |
| "logits/chosen": -0.5727726221084595, | |
| "logits/rejected": -1.5278774499893188, | |
| "logps/chosen": -340.2919921875, | |
| "logps/rejected": -246.67628479003906, | |
| "loss": 0.0183, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.279099464416504, | |
| "rewards/margins": 23.8295841217041, | |
| "rewards/rejected": -16.550485610961914, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.18448275862069e-05, | |
| "logits/chosen": -0.588262677192688, | |
| "logits/rejected": -1.4113011360168457, | |
| "logps/chosen": -239.1262664794922, | |
| "logps/rejected": -323.8817443847656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.26235294342041, | |
| "rewards/margins": 25.60938262939453, | |
| "rewards/rejected": -21.34703254699707, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.179310344827587e-05, | |
| "logits/chosen": -0.825312077999115, | |
| "logits/rejected": -1.6536335945129395, | |
| "logps/chosen": -245.26333618164062, | |
| "logps/rejected": -275.9490966796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.488056182861328, | |
| "rewards/margins": 24.79592514038086, | |
| "rewards/rejected": -20.30786895751953, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.174137931034483e-05, | |
| "logits/chosen": -0.6509658694267273, | |
| "logits/rejected": -1.7768073081970215, | |
| "logps/chosen": -494.1702575683594, | |
| "logps/rejected": -269.7856140136719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 11.041641235351562, | |
| "rewards/margins": 30.055639266967773, | |
| "rewards/rejected": -19.013999938964844, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.16896551724138e-05, | |
| "logits/chosen": -0.813829243183136, | |
| "logits/rejected": -1.7704083919525146, | |
| "logps/chosen": -370.6927795410156, | |
| "logps/rejected": -288.31463623046875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.1367826461792, | |
| "rewards/margins": 30.452945709228516, | |
| "rewards/rejected": -20.316162109375, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.163793103448276e-05, | |
| "logits/chosen": -0.450665146112442, | |
| "logits/rejected": -1.4787647724151611, | |
| "logps/chosen": -336.4142761230469, | |
| "logps/rejected": -275.8489990234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.380362510681152, | |
| "rewards/margins": 27.303016662597656, | |
| "rewards/rejected": -20.922651290893555, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.158620689655173e-05, | |
| "logits/chosen": -0.7610673904418945, | |
| "logits/rejected": -1.5079166889190674, | |
| "logps/chosen": -290.8649597167969, | |
| "logps/rejected": -391.47772216796875, | |
| "loss": 0.4671, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 4.447336196899414, | |
| "rewards/margins": 27.807403564453125, | |
| "rewards/rejected": -23.36006736755371, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.153448275862069e-05, | |
| "logits/chosen": -0.9873843193054199, | |
| "logits/rejected": -1.6280449628829956, | |
| "logps/chosen": -244.6575164794922, | |
| "logps/rejected": -324.1699523925781, | |
| "loss": 0.0036, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1840460300445557, | |
| "rewards/margins": 25.491558074951172, | |
| "rewards/rejected": -23.307514190673828, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.148275862068966e-05, | |
| "logits/chosen": -1.0418487787246704, | |
| "logits/rejected": -1.7689330577850342, | |
| "logps/chosen": -309.9642028808594, | |
| "logps/rejected": -433.8427734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.8272902965545654, | |
| "rewards/margins": 33.52722930908203, | |
| "rewards/rejected": -29.69993782043457, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.143103448275862e-05, | |
| "logits/chosen": -0.9574589133262634, | |
| "logits/rejected": -1.9435979127883911, | |
| "logps/chosen": -485.497314453125, | |
| "logps/rejected": -301.46038818359375, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.4343650341033936, | |
| "rewards/margins": 27.89614486694336, | |
| "rewards/rejected": -24.46177864074707, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "logits/chosen": -1.2192739248275757, | |
| "logits/rejected": -1.9555035829544067, | |
| "logps/chosen": -485.7822265625, | |
| "logps/rejected": -452.36566162109375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4936442375183105, | |
| "rewards/margins": 31.84035873413086, | |
| "rewards/rejected": -35.33400344848633, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.1327586206896555e-05, | |
| "logits/chosen": -1.3901960849761963, | |
| "logits/rejected": -2.194340944290161, | |
| "logps/chosen": -556.4442749023438, | |
| "logps/rejected": -403.41094970703125, | |
| "loss": 0.4842, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -6.604147434234619, | |
| "rewards/margins": 25.96053695678711, | |
| "rewards/rejected": -32.5646858215332, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.1275862068965516e-05, | |
| "logits/chosen": -1.10233473777771, | |
| "logits/rejected": -1.911782145500183, | |
| "logps/chosen": -319.7232666015625, | |
| "logps/rejected": -366.8639221191406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.5800843238830566, | |
| "rewards/margins": 30.92066192626953, | |
| "rewards/rejected": -30.3405818939209, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.1224137931034484e-05, | |
| "logits/chosen": -0.7030006051063538, | |
| "logits/rejected": -1.7387290000915527, | |
| "logps/chosen": -291.53961181640625, | |
| "logps/rejected": -350.3901062011719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.4014167785644531, | |
| "rewards/margins": 28.830717086791992, | |
| "rewards/rejected": -27.429298400878906, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.117241379310345e-05, | |
| "logits/chosen": -0.5296112895011902, | |
| "logits/rejected": -1.5072706937789917, | |
| "logps/chosen": -286.94879150390625, | |
| "logps/rejected": -346.3030700683594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.8979191780090332, | |
| "rewards/margins": 28.363784790039062, | |
| "rewards/rejected": -26.465864181518555, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.112068965517242e-05, | |
| "logits/chosen": -0.42470210790634155, | |
| "logits/rejected": -1.7659740447998047, | |
| "logps/chosen": -477.8876953125, | |
| "logps/rejected": -315.16717529296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.757699966430664, | |
| "rewards/margins": 34.151214599609375, | |
| "rewards/rejected": -24.393516540527344, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.106896551724138e-05, | |
| "logits/chosen": -0.37895911931991577, | |
| "logits/rejected": -1.4872817993164062, | |
| "logps/chosen": -360.4671936035156, | |
| "logps/rejected": -311.3260803222656, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.024797439575195, | |
| "rewards/margins": 30.34416961669922, | |
| "rewards/rejected": -20.31937026977539, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.101724137931035e-05, | |
| "logits/chosen": -0.4127517342567444, | |
| "logits/rejected": -1.4179290533065796, | |
| "logps/chosen": -396.7817077636719, | |
| "logps/rejected": -257.0691833496094, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.115121841430664, | |
| "rewards/margins": 25.882369995117188, | |
| "rewards/rejected": -17.767250061035156, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.096551724137932e-05, | |
| "logits/chosen": -0.4867606461048126, | |
| "logits/rejected": -1.433240532875061, | |
| "logps/chosen": -307.6831359863281, | |
| "logps/rejected": -285.9851379394531, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.156608581542969, | |
| "rewards/margins": 24.28235626220703, | |
| "rewards/rejected": -15.12574577331543, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.091379310344828e-05, | |
| "logits/chosen": -0.5465874671936035, | |
| "logits/rejected": -1.398258924484253, | |
| "logps/chosen": -244.0363311767578, | |
| "logps/rejected": -240.68551635742188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.43398904800415, | |
| "rewards/margins": 23.1319522857666, | |
| "rewards/rejected": -18.69796371459961, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.0862068965517246e-05, | |
| "logits/chosen": -0.38893836736679077, | |
| "logits/rejected": -1.427646279335022, | |
| "logps/chosen": -351.76348876953125, | |
| "logps/rejected": -289.0392150878906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.574392795562744, | |
| "rewards/margins": 25.838218688964844, | |
| "rewards/rejected": -18.26382827758789, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.0810344827586214e-05, | |
| "logits/chosen": -0.3710274398326874, | |
| "logits/rejected": -1.5999269485473633, | |
| "logps/chosen": -394.9821472167969, | |
| "logps/rejected": -246.48887634277344, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.973166465759277, | |
| "rewards/margins": 26.636192321777344, | |
| "rewards/rejected": -17.66302490234375, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.0758620689655175e-05, | |
| "logits/chosen": -0.673173725605011, | |
| "logits/rejected": -1.2474948167800903, | |
| "logps/chosen": -257.3236083984375, | |
| "logps/rejected": -264.7505187988281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.773941993713379, | |
| "rewards/margins": 19.319332122802734, | |
| "rewards/rejected": -16.545391082763672, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.0706896551724143e-05, | |
| "logits/chosen": -0.560947060585022, | |
| "logits/rejected": -1.374908447265625, | |
| "logps/chosen": -301.8670959472656, | |
| "logps/rejected": -232.6251983642578, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.369309425354004, | |
| "rewards/margins": 26.155746459960938, | |
| "rewards/rejected": -15.786436080932617, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.0655172413793105e-05, | |
| "logits/chosen": -0.6484676003456116, | |
| "logits/rejected": -1.3061511516571045, | |
| "logps/chosen": -240.1058807373047, | |
| "logps/rejected": -292.108154296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.469289302825928, | |
| "rewards/margins": 25.45884132385254, | |
| "rewards/rejected": -20.989551544189453, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.060344827586207e-05, | |
| "logits/chosen": -0.4261731505393982, | |
| "logits/rejected": -1.5151227712631226, | |
| "logps/chosen": -402.2912292480469, | |
| "logps/rejected": -289.1283264160156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.882993698120117, | |
| "rewards/margins": 30.517051696777344, | |
| "rewards/rejected": -19.63405990600586, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.0551724137931034e-05, | |
| "logits/chosen": -0.3702995181083679, | |
| "logits/rejected": -1.5198004245758057, | |
| "logps/chosen": -377.6826477050781, | |
| "logps/rejected": -287.36566162109375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.187080383300781, | |
| "rewards/margins": 30.941383361816406, | |
| "rewards/rejected": -20.754301071166992, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.05e-05, | |
| "logits/chosen": -0.394854873418808, | |
| "logits/rejected": -1.641805648803711, | |
| "logps/chosen": -462.9255676269531, | |
| "logps/rejected": -254.52276611328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.504643440246582, | |
| "rewards/margins": 27.564743041992188, | |
| "rewards/rejected": -17.060100555419922, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.044827586206896e-05, | |
| "logits/chosen": -0.3886590600013733, | |
| "logits/rejected": -1.4597467184066772, | |
| "logps/chosen": -378.5803527832031, | |
| "logps/rejected": -263.2305908203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 13.531683921813965, | |
| "rewards/margins": 31.480838775634766, | |
| "rewards/rejected": -17.949153900146484, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.039655172413793e-05, | |
| "logits/chosen": -0.5102599859237671, | |
| "logits/rejected": -1.250558614730835, | |
| "logps/chosen": -295.6455078125, | |
| "logps/rejected": -345.3118591308594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.247247695922852, | |
| "rewards/margins": 26.292499542236328, | |
| "rewards/rejected": -19.04525375366211, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.03448275862069e-05, | |
| "logits/chosen": -0.39377278089523315, | |
| "logits/rejected": -1.5313684940338135, | |
| "logps/chosen": -441.0325622558594, | |
| "logps/rejected": -223.12750244140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 12.754582405090332, | |
| "rewards/margins": 28.21717071533203, | |
| "rewards/rejected": -15.462587356567383, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.029310344827587e-05, | |
| "logits/chosen": -0.45350462198257446, | |
| "logits/rejected": -1.3594186305999756, | |
| "logps/chosen": -313.7057189941406, | |
| "logps/rejected": -304.0419616699219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.3996429443359375, | |
| "rewards/margins": 23.64727020263672, | |
| "rewards/rejected": -17.24762725830078, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.024137931034483e-05, | |
| "logits/chosen": -0.17602750658988953, | |
| "logits/rejected": -1.2745662927627563, | |
| "logps/chosen": -371.0513916015625, | |
| "logps/rejected": -299.74481201171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.569783687591553, | |
| "rewards/margins": 26.724822998046875, | |
| "rewards/rejected": -19.155040740966797, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.0189655172413796e-05, | |
| "logits/chosen": -0.44520992040634155, | |
| "logits/rejected": -1.4310128688812256, | |
| "logps/chosen": -334.8761901855469, | |
| "logps/rejected": -278.89349365234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.9243433475494385, | |
| "rewards/margins": 22.313913345336914, | |
| "rewards/rejected": -18.389568328857422, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.0137931034482764e-05, | |
| "logits/chosen": -0.3496861755847931, | |
| "logits/rejected": -1.480123519897461, | |
| "logps/chosen": -345.8548583984375, | |
| "logps/rejected": -295.12042236328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.938793182373047, | |
| "rewards/margins": 28.614458084106445, | |
| "rewards/rejected": -18.67566680908203, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.0086206896551726e-05, | |
| "logits/chosen": -0.4202902913093567, | |
| "logits/rejected": -1.5481184720993042, | |
| "logps/chosen": -408.18121337890625, | |
| "logps/rejected": -273.8868408203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.119111061096191, | |
| "rewards/margins": 29.09032440185547, | |
| "rewards/rejected": -19.971214294433594, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.0034482758620694e-05, | |
| "logits/chosen": -0.5120177268981934, | |
| "logits/rejected": -1.4507927894592285, | |
| "logps/chosen": -271.6726989746094, | |
| "logps/rejected": -297.4678955078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.688560962677002, | |
| "rewards/margins": 26.492656707763672, | |
| "rewards/rejected": -20.804094314575195, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.998275862068966e-05, | |
| "logits/chosen": -0.41148996353149414, | |
| "logits/rejected": -1.4985872507095337, | |
| "logps/chosen": -343.8206481933594, | |
| "logps/rejected": -241.0497589111328, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.314858436584473, | |
| "rewards/margins": 26.18862533569336, | |
| "rewards/rejected": -16.87376594543457, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.993103448275862e-05, | |
| "logits/chosen": -0.1992441862821579, | |
| "logits/rejected": -1.8458625078201294, | |
| "logps/chosen": -504.0348205566406, | |
| "logps/rejected": -318.8588562011719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 12.68671989440918, | |
| "rewards/margins": 34.4812126159668, | |
| "rewards/rejected": -21.79449462890625, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.987931034482759e-05, | |
| "logits/chosen": -0.5546131730079651, | |
| "logits/rejected": -1.527896761894226, | |
| "logps/chosen": -333.9448547363281, | |
| "logps/rejected": -303.9145202636719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 10.168367385864258, | |
| "rewards/margins": 30.0524959564209, | |
| "rewards/rejected": -19.88412857055664, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.982758620689656e-05, | |
| "logits/chosen": -0.3886173367500305, | |
| "logits/rejected": -1.621660828590393, | |
| "logps/chosen": -369.8122253417969, | |
| "logps/rejected": -272.4569396972656, | |
| "loss": 0.0016, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.674747467041016, | |
| "rewards/margins": 26.678119659423828, | |
| "rewards/rejected": -18.003372192382812, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.977586206896552e-05, | |
| "logits/chosen": -0.61821049451828, | |
| "logits/rejected": -1.6792049407958984, | |
| "logps/chosen": -345.2456359863281, | |
| "logps/rejected": -307.6291809082031, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.595407485961914, | |
| "rewards/margins": 29.619413375854492, | |
| "rewards/rejected": -22.024009704589844, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.972413793103449e-05, | |
| "logits/chosen": -0.23747998476028442, | |
| "logits/rejected": -1.287044644355774, | |
| "logps/chosen": -361.4759521484375, | |
| "logps/rejected": -267.63592529296875, | |
| "loss": 0.2105, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 7.393239974975586, | |
| "rewards/margins": 23.982959747314453, | |
| "rewards/rejected": -16.5897216796875, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.967241379310345e-05, | |
| "logits/chosen": -0.3899969160556793, | |
| "logits/rejected": -1.6496773958206177, | |
| "logps/chosen": -331.9876403808594, | |
| "logps/rejected": -278.04522705078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.100687026977539, | |
| "rewards/margins": 29.64002227783203, | |
| "rewards/rejected": -21.539335250854492, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.962068965517242e-05, | |
| "logits/chosen": -0.4621184766292572, | |
| "logits/rejected": -1.3823509216308594, | |
| "logps/chosen": -291.25494384765625, | |
| "logps/rejected": -313.21734619140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.036273002624512, | |
| "rewards/margins": 25.113258361816406, | |
| "rewards/rejected": -21.076988220214844, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.956896551724138e-05, | |
| "logits/chosen": -0.3050623834133148, | |
| "logits/rejected": -1.3223061561584473, | |
| "logps/chosen": -388.243896484375, | |
| "logps/rejected": -266.01983642578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.008649826049805, | |
| "rewards/margins": 24.918262481689453, | |
| "rewards/rejected": -17.90961265563965, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.9517241379310346e-05, | |
| "logits/chosen": -0.26263201236724854, | |
| "logits/rejected": -1.262205958366394, | |
| "logps/chosen": -300.5107421875, | |
| "logps/rejected": -266.810302734375, | |
| "loss": 0.0007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.9579243659973145, | |
| "rewards/margins": 23.850536346435547, | |
| "rewards/rejected": -17.892614364624023, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.9465517241379314e-05, | |
| "logits/chosen": -0.34424301981925964, | |
| "logits/rejected": -1.2062197923660278, | |
| "logps/chosen": -273.7149353027344, | |
| "logps/rejected": -266.54150390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.983701705932617, | |
| "rewards/margins": 26.13591957092285, | |
| "rewards/rejected": -20.1522216796875, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.9413793103448276e-05, | |
| "logits/chosen": -0.1809622049331665, | |
| "logits/rejected": -1.4350645542144775, | |
| "logps/chosen": -418.6148376464844, | |
| "logps/rejected": -297.8706359863281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 8.36530876159668, | |
| "rewards/margins": 30.20969581604004, | |
| "rewards/rejected": -21.84438705444336, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.9362068965517244e-05, | |
| "logits/chosen": -0.3474291265010834, | |
| "logits/rejected": -1.3261463642120361, | |
| "logps/chosen": -392.76068115234375, | |
| "logps/rejected": -282.137451171875, | |
| "loss": 0.12, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 8.480328559875488, | |
| "rewards/margins": 24.708953857421875, | |
| "rewards/rejected": -16.228626251220703, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.931034482758621e-05, | |
| "logits/chosen": -0.29881787300109863, | |
| "logits/rejected": -1.59579598903656, | |
| "logps/chosen": -491.68280029296875, | |
| "logps/rejected": -329.0652160644531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 9.562919616699219, | |
| "rewards/margins": 31.969928741455078, | |
| "rewards/rejected": -22.407007217407227, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.925862068965517e-05, | |
| "logits/chosen": -0.24624793231487274, | |
| "logits/rejected": -1.2760369777679443, | |
| "logps/chosen": -309.42950439453125, | |
| "logps/rejected": -337.400634765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.126012802124023, | |
| "rewards/margins": 27.284088134765625, | |
| "rewards/rejected": -22.1580753326416, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.920689655172414e-05, | |
| "logits/chosen": -0.24486477673053741, | |
| "logits/rejected": -1.4145996570587158, | |
| "logps/chosen": -435.2896423339844, | |
| "logps/rejected": -347.84698486328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.421405076980591, | |
| "rewards/margins": 30.74762725830078, | |
| "rewards/rejected": -27.32622528076172, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.915517241379311e-05, | |
| "logits/chosen": -0.4706140458583832, | |
| "logits/rejected": -1.4215468168258667, | |
| "logps/chosen": -427.6326904296875, | |
| "logps/rejected": -399.431640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.1375885009765625, | |
| "rewards/margins": 31.478370666503906, | |
| "rewards/rejected": -27.340782165527344, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.910344827586207e-05, | |
| "logits/chosen": -0.25800830125808716, | |
| "logits/rejected": -1.604130744934082, | |
| "logps/chosen": -430.5594482421875, | |
| "logps/rejected": -431.40301513671875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.8110466003417969, | |
| "rewards/margins": 36.45166015625, | |
| "rewards/rejected": -34.64060974121094, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.905172413793104e-05, | |
| "logits/chosen": -0.3043404221534729, | |
| "logits/rejected": -1.5278654098510742, | |
| "logps/chosen": -465.7344970703125, | |
| "logps/rejected": -436.65899658203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.1547170877456665, | |
| "rewards/margins": 33.50754928588867, | |
| "rewards/rejected": -34.66226577758789, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "logits/chosen": -0.22293318808078766, | |
| "logits/rejected": -1.2031610012054443, | |
| "logps/chosen": -428.4457092285156, | |
| "logps/rejected": -419.1228332519531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.0903587341308594, | |
| "rewards/margins": 28.546253204345703, | |
| "rewards/rejected": -31.636613845825195, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.894827586206897e-05, | |
| "logits/chosen": -0.18892326951026917, | |
| "logits/rejected": -1.481589913368225, | |
| "logps/chosen": -513.7359008789062, | |
| "logps/rejected": -456.95208740234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.6842398643493652, | |
| "rewards/margins": 39.1934928894043, | |
| "rewards/rejected": -36.509254455566406, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.8896551724137935e-05, | |
| "logits/chosen": -0.48873329162597656, | |
| "logits/rejected": -1.4820678234100342, | |
| "logps/chosen": -393.00079345703125, | |
| "logps/rejected": -426.32696533203125, | |
| "loss": 0.0194, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.972713470458984, | |
| "rewards/margins": 28.33656120300293, | |
| "rewards/rejected": -34.30927276611328, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.88448275862069e-05, | |
| "logits/chosen": -0.4431387484073639, | |
| "logits/rejected": -1.3382742404937744, | |
| "logps/chosen": -392.2378845214844, | |
| "logps/rejected": -503.30194091796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.949856758117676, | |
| "rewards/margins": 34.904537200927734, | |
| "rewards/rejected": -40.854393005371094, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.8793103448275865e-05, | |
| "logits/chosen": -0.2996464669704437, | |
| "logits/rejected": -1.4847010374069214, | |
| "logps/chosen": -483.34161376953125, | |
| "logps/rejected": -548.205078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.7714486122131348, | |
| "rewards/margins": 47.965972900390625, | |
| "rewards/rejected": -44.19452667236328, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.874137931034483e-05, | |
| "logits/chosen": -0.3217467963695526, | |
| "logits/rejected": -1.2237169742584229, | |
| "logps/chosen": -396.9007568359375, | |
| "logps/rejected": -509.6751708984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.9992933869361877, | |
| "rewards/margins": 39.428592681884766, | |
| "rewards/rejected": -38.429298400878906, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.8689655172413794e-05, | |
| "logits/chosen": -0.26858553290367126, | |
| "logits/rejected": -1.3162765502929688, | |
| "logps/chosen": -436.47088623046875, | |
| "logps/rejected": -461.6593933105469, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.6044464111328125, | |
| "rewards/margins": 38.4824104309082, | |
| "rewards/rejected": -35.87796401977539, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.863793103448276e-05, | |
| "logits/chosen": -0.343127965927124, | |
| "logits/rejected": -1.390782117843628, | |
| "logps/chosen": -392.77978515625, | |
| "logps/rejected": -443.3890686035156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.19043821096420288, | |
| "rewards/margins": 34.40040969848633, | |
| "rewards/rejected": -34.20996856689453, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.858620689655172e-05, | |
| "logits/chosen": -0.43945077061653137, | |
| "logits/rejected": -1.3458408117294312, | |
| "logps/chosen": -386.0972595214844, | |
| "logps/rejected": -481.8536071777344, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.2730538845062256, | |
| "rewards/margins": 35.0412483215332, | |
| "rewards/rejected": -33.76819610595703, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.853448275862069e-05, | |
| "logits/chosen": -0.28766676783561707, | |
| "logits/rejected": -1.314267873764038, | |
| "logps/chosen": -423.24981689453125, | |
| "logps/rejected": -416.8682861328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.9727609157562256, | |
| "rewards/margins": 34.98452377319336, | |
| "rewards/rejected": -33.01176834106445, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.848275862068966e-05, | |
| "logits/chosen": -0.3893183767795563, | |
| "logits/rejected": -1.443713903427124, | |
| "logps/chosen": -474.935546875, | |
| "logps/rejected": -453.4057922363281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1096689701080322, | |
| "rewards/margins": 38.80006790161133, | |
| "rewards/rejected": -36.690399169921875, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.843103448275862e-05, | |
| "logits/chosen": -0.34813588857650757, | |
| "logits/rejected": -1.271755337715149, | |
| "logps/chosen": -392.284423828125, | |
| "logps/rejected": -496.442138671875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.8468620777130127, | |
| "rewards/margins": 41.135921478271484, | |
| "rewards/rejected": -38.289058685302734, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.837931034482759e-05, | |
| "logits/chosen": -0.3464352786540985, | |
| "logits/rejected": -1.223905324935913, | |
| "logps/chosen": -372.1242370605469, | |
| "logps/rejected": -446.0730285644531, | |
| "loss": 0.0073, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.18368953466415405, | |
| "rewards/margins": 33.218013763427734, | |
| "rewards/rejected": -33.034324645996094, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.8327586206896556e-05, | |
| "logits/chosen": -0.42792245745658875, | |
| "logits/rejected": -1.189011812210083, | |
| "logps/chosen": -324.2301940917969, | |
| "logps/rejected": -503.5957946777344, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.7187633514404297, | |
| "rewards/margins": 37.19697189331055, | |
| "rewards/rejected": -38.91573715209961, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.827586206896552e-05, | |
| "logits/chosen": -0.16668730974197388, | |
| "logits/rejected": -1.5985019207000732, | |
| "logps/chosen": -482.8082580566406, | |
| "logps/rejected": -449.761474609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.7022485733032227, | |
| "rewards/margins": 42.01506805419922, | |
| "rewards/rejected": -38.31282043457031, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.8224137931034485e-05, | |
| "logits/chosen": -0.26949506998062134, | |
| "logits/rejected": -1.2986763715744019, | |
| "logps/chosen": -348.618896484375, | |
| "logps/rejected": -442.8262023925781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1358022689819336, | |
| "rewards/margins": 35.771461486816406, | |
| "rewards/rejected": -33.63566207885742, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.8172413793103453e-05, | |
| "logits/chosen": -0.3438015580177307, | |
| "logits/rejected": -1.3001600503921509, | |
| "logps/chosen": -333.9195251464844, | |
| "logps/rejected": -457.9849853515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.145051956176758, | |
| "rewards/margins": 41.207923889160156, | |
| "rewards/rejected": -39.06287384033203, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.8120689655172415e-05, | |
| "logits/chosen": -0.10503964871168137, | |
| "logits/rejected": -1.2384865283966064, | |
| "logps/chosen": -451.48663330078125, | |
| "logps/rejected": -479.8037414550781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.525763988494873, | |
| "rewards/margins": 38.3972282409668, | |
| "rewards/rejected": -37.8714599609375, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.806896551724138e-05, | |
| "logits/chosen": -0.30969691276550293, | |
| "logits/rejected": -1.1278965473175049, | |
| "logps/chosen": -298.7955322265625, | |
| "logps/rejected": -499.42279052734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.1820168495178223, | |
| "rewards/margins": 37.28181457519531, | |
| "rewards/rejected": -39.463829040527344, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.801724137931035e-05, | |
| "logits/chosen": -0.3709387183189392, | |
| "logits/rejected": -1.5942573547363281, | |
| "logps/chosen": -422.3631286621094, | |
| "logps/rejected": -511.52008056640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.8516347408294678, | |
| "rewards/margins": 41.39206314086914, | |
| "rewards/rejected": -38.540428161621094, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.796551724137931e-05, | |
| "logits/chosen": -0.2693447470664978, | |
| "logits/rejected": -1.5490005016326904, | |
| "logps/chosen": -480.881591796875, | |
| "logps/rejected": -414.71917724609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.7573628425598145, | |
| "rewards/margins": 39.35446548461914, | |
| "rewards/rejected": -36.59709930419922, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.791379310344828e-05, | |
| "logits/chosen": -0.39544716477394104, | |
| "logits/rejected": -1.4232298135757446, | |
| "logps/chosen": -400.05267333984375, | |
| "logps/rejected": -465.1136169433594, | |
| "loss": 0.0018, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.5475125312805176, | |
| "rewards/margins": 38.094913482666016, | |
| "rewards/rejected": -34.547393798828125, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.786206896551725e-05, | |
| "logits/chosen": -0.3474569320678711, | |
| "logits/rejected": -1.4260185956954956, | |
| "logps/chosen": -334.5423889160156, | |
| "logps/rejected": -438.87408447265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.5236696004867554, | |
| "rewards/margins": 33.86663055419922, | |
| "rewards/rejected": -35.390296936035156, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.781034482758621e-05, | |
| "logits/chosen": -0.3836357593536377, | |
| "logits/rejected": -1.554406762123108, | |
| "logps/chosen": -372.57830810546875, | |
| "logps/rejected": -439.4316101074219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.840342402458191, | |
| "rewards/margins": 34.46831512451172, | |
| "rewards/rejected": -32.627967834472656, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.775862068965517e-05, | |
| "logits/chosen": -0.18503640592098236, | |
| "logits/rejected": -1.800915241241455, | |
| "logps/chosen": -485.19354248046875, | |
| "logps/rejected": -432.1221923828125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.203826427459717, | |
| "rewards/margins": 39.722900390625, | |
| "rewards/rejected": -35.51907730102539, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.770689655172414e-05, | |
| "logits/chosen": -0.3300974667072296, | |
| "logits/rejected": -1.5052634477615356, | |
| "logps/chosen": -418.87017822265625, | |
| "logps/rejected": -408.99554443359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9931721687316895, | |
| "rewards/margins": 35.2595100402832, | |
| "rewards/rejected": -32.266334533691406, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.7655172413793106e-05, | |
| "logits/chosen": -0.27791130542755127, | |
| "logits/rejected": -1.4637497663497925, | |
| "logps/chosen": -370.642822265625, | |
| "logps/rejected": -457.7041320800781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.304169178009033, | |
| "rewards/margins": 40.587310791015625, | |
| "rewards/rejected": -36.283138275146484, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.760344827586207e-05, | |
| "logits/chosen": -0.4701173007488251, | |
| "logits/rejected": -1.403714895248413, | |
| "logps/chosen": -333.7972717285156, | |
| "logps/rejected": -436.09967041015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.8245915174484253, | |
| "rewards/margins": 36.37276077270508, | |
| "rewards/rejected": -34.54816818237305, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.7551724137931035e-05, | |
| "logits/chosen": -0.08819374442100525, | |
| "logits/rejected": -1.395701289176941, | |
| "logps/chosen": -401.96673583984375, | |
| "logps/rejected": -440.0989685058594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.565300941467285, | |
| "rewards/margins": 33.236175537109375, | |
| "rewards/rejected": -30.670875549316406, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "logits/chosen": -0.17811371386051178, | |
| "logits/rejected": -1.501448154449463, | |
| "logps/chosen": -447.5896301269531, | |
| "logps/rejected": -375.3368835449219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.7833425998687744, | |
| "rewards/margins": 31.601383209228516, | |
| "rewards/rejected": -28.818038940429688, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.7448275862068965e-05, | |
| "logits/chosen": -0.44955122470855713, | |
| "logits/rejected": -1.550938606262207, | |
| "logps/chosen": -289.89959716796875, | |
| "logps/rejected": -400.880859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.4812740385532379, | |
| "rewards/margins": 33.061256408691406, | |
| "rewards/rejected": -33.54253005981445, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.739655172413793e-05, | |
| "logits/chosen": -0.18104656040668488, | |
| "logits/rejected": -1.4971003532409668, | |
| "logps/chosen": -418.8290710449219, | |
| "logps/rejected": -424.74859619140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.926714897155762, | |
| "rewards/margins": 39.01382064819336, | |
| "rewards/rejected": -32.08710861206055, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.73448275862069e-05, | |
| "logits/chosen": -0.16005319356918335, | |
| "logits/rejected": -1.4771628379821777, | |
| "logps/chosen": -437.9535827636719, | |
| "logps/rejected": -435.7216796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.183934688568115, | |
| "rewards/margins": 40.51701354980469, | |
| "rewards/rejected": -35.33308410644531, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.729310344827586e-05, | |
| "logits/chosen": -0.35886967182159424, | |
| "logits/rejected": -1.6980268955230713, | |
| "logps/chosen": -404.4248962402344, | |
| "logps/rejected": -412.4001159667969, | |
| "loss": 0.0068, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 7.873406887054443, | |
| "rewards/margins": 38.6983757019043, | |
| "rewards/rejected": -30.824968338012695, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.724137931034483e-05, | |
| "logits/chosen": -0.4241984486579895, | |
| "logits/rejected": -1.4724136590957642, | |
| "logps/chosen": -353.877197265625, | |
| "logps/rejected": -414.2342529296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.698596954345703, | |
| "rewards/margins": 36.431583404541016, | |
| "rewards/rejected": -33.73298645019531, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.71896551724138e-05, | |
| "logits/chosen": -0.38190075755119324, | |
| "logits/rejected": -1.520878553390503, | |
| "logps/chosen": -303.9383544921875, | |
| "logps/rejected": -443.4150390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 5.500533103942871, | |
| "rewards/margins": 41.76572036743164, | |
| "rewards/rejected": -36.26518630981445, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.713793103448276e-05, | |
| "logits/chosen": -0.5008329153060913, | |
| "logits/rejected": -1.4159965515136719, | |
| "logps/chosen": -298.4696044921875, | |
| "logps/rejected": -496.4792175292969, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.6865413188934326, | |
| "rewards/margins": 43.41154098510742, | |
| "rewards/rejected": -39.72499465942383, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.708620689655173e-05, | |
| "logits/chosen": -0.30573636293411255, | |
| "logits/rejected": -1.4757778644561768, | |
| "logps/chosen": -404.08349609375, | |
| "logps/rejected": -536.4090576171875, | |
| "loss": 0.0994, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": 2.837350606918335, | |
| "rewards/margins": 47.625877380371094, | |
| "rewards/rejected": -44.78852844238281, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.7034482758620695e-05, | |
| "logits/chosen": -0.08167792111635208, | |
| "logits/rejected": -1.6543066501617432, | |
| "logps/chosen": -478.5865173339844, | |
| "logps/rejected": -535.1455078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 6.218913555145264, | |
| "rewards/margins": 51.752662658691406, | |
| "rewards/rejected": -45.533748626708984, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 916, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |