| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 302, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.0, | |
| "logits/chosen": -4.7122321128845215, | |
| "logits/rejected": -3.4910330772399902, | |
| "logps/chosen": -480.7637939453125, | |
| "logps/rejected": -307.30804443359375, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0, | |
| "logits/chosen": -4.456130027770996, | |
| "logits/rejected": -3.5022683143615723, | |
| "logps/chosen": -441.0960693359375, | |
| "logps/rejected": -254.25729370117188, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0, | |
| "logits/chosen": -4.721767425537109, | |
| "logits/rejected": -3.4279396533966064, | |
| "logps/chosen": -440.868896484375, | |
| "logps/rejected": -285.726806640625, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5.494505494505495e-08, | |
| "logits/chosen": -4.769167423248291, | |
| "logits/rejected": -4.213985443115234, | |
| "logps/chosen": -339.8769226074219, | |
| "logps/rejected": -288.4583435058594, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.098901098901099e-07, | |
| "logits/chosen": -4.547554016113281, | |
| "logits/rejected": -3.9193525314331055, | |
| "logps/chosen": -351.4005126953125, | |
| "logps/rejected": -348.05291748046875, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.6483516483516484e-07, | |
| "logits/chosen": -4.506200790405273, | |
| "logits/rejected": -4.258366107940674, | |
| "logps/chosen": -367.66943359375, | |
| "logps/rejected": -365.24383544921875, | |
| "loss": 0.701, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": -0.00408935546875, | |
| "rewards/margins": -0.01570434495806694, | |
| "rewards/rejected": 0.011614990420639515, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.197802197802198e-07, | |
| "logits/chosen": -4.512852191925049, | |
| "logits/rejected": -3.491598129272461, | |
| "logps/chosen": -418.4132995605469, | |
| "logps/rejected": -340.18426513671875, | |
| "loss": 0.6811, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.0052337646484375, | |
| "rewards/margins": 0.0241851806640625, | |
| "rewards/rejected": -0.018951416015625, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.7472527472527475e-07, | |
| "logits/chosen": -4.398182392120361, | |
| "logits/rejected": -3.73840069770813, | |
| "logps/chosen": -383.665771484375, | |
| "logps/rejected": -369.041748046875, | |
| "loss": 0.6499, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.03215942531824112, | |
| "rewards/margins": 0.088531494140625, | |
| "rewards/rejected": -0.05637207254767418, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.296703296703297e-07, | |
| "logits/chosen": -4.551479339599609, | |
| "logits/rejected": -3.8382506370544434, | |
| "logps/chosen": -348.11627197265625, | |
| "logps/rejected": -271.0714111328125, | |
| "loss": 0.6261, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.06197509914636612, | |
| "rewards/margins": 0.13901062309741974, | |
| "rewards/rejected": -0.07703552395105362, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.846153846153847e-07, | |
| "logits/chosen": -4.554244518280029, | |
| "logits/rejected": -3.5874130725860596, | |
| "logps/chosen": -477.6059875488281, | |
| "logps/rejected": -331.3657531738281, | |
| "loss": 0.5562, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.08335571736097336, | |
| "rewards/margins": 0.2956085205078125, | |
| "rewards/rejected": -0.21225281059741974, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.395604395604396e-07, | |
| "logits/chosen": -4.534104347229004, | |
| "logits/rejected": -3.622457265853882, | |
| "logps/chosen": -452.00927734375, | |
| "logps/rejected": -308.62396240234375, | |
| "loss": 0.5024, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.11546631157398224, | |
| "rewards/margins": 0.42655640840530396, | |
| "rewards/rejected": -0.3110900819301605, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.945054945054946e-07, | |
| "logits/chosen": -4.5819878578186035, | |
| "logits/rejected": -4.043532848358154, | |
| "logps/chosen": -489.98583984375, | |
| "logps/rejected": -355.9280700683594, | |
| "loss": 0.3914, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.254067987203598, | |
| "rewards/margins": 0.7358551025390625, | |
| "rewards/rejected": -0.4817871153354645, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.494505494505495e-07, | |
| "logits/chosen": -4.548617839813232, | |
| "logits/rejected": -3.5121090412139893, | |
| "logps/chosen": -453.76708984375, | |
| "logps/rejected": -340.49310302734375, | |
| "loss": 0.3846, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.16141052544116974, | |
| "rewards/margins": 0.7571258544921875, | |
| "rewards/rejected": -0.595715343952179, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.043956043956044e-07, | |
| "logits/chosen": -4.559290885925293, | |
| "logits/rejected": -3.851238250732422, | |
| "logps/chosen": -359.08013916015625, | |
| "logps/rejected": -325.69927978515625, | |
| "loss": 0.2665, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.43179628252983093, | |
| "rewards/margins": 1.1861114501953125, | |
| "rewards/rejected": -0.754315197467804, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 6.593406593406594e-07, | |
| "logits/chosen": -4.530972003936768, | |
| "logits/rejected": -3.6676721572875977, | |
| "logps/chosen": -283.534423828125, | |
| "logps/rejected": -344.08294677734375, | |
| "loss": 0.2007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.46563720703125, | |
| "rewards/margins": 1.5038880109786987, | |
| "rewards/rejected": -1.0382508039474487, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 7.142857142857143e-07, | |
| "logits/chosen": -4.578598976135254, | |
| "logits/rejected": -3.969371795654297, | |
| "logps/chosen": -329.626708984375, | |
| "logps/rejected": -358.00067138671875, | |
| "loss": 0.15, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.37066346406936646, | |
| "rewards/margins": 1.8208892345428467, | |
| "rewards/rejected": -1.450225830078125, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 7.692307692307694e-07, | |
| "logits/chosen": -4.456603050231934, | |
| "logits/rejected": -3.9640321731567383, | |
| "logps/chosen": -325.3027648925781, | |
| "logps/rejected": -371.80010986328125, | |
| "loss": 0.2361, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.45557862520217896, | |
| "rewards/margins": 1.3231537342071533, | |
| "rewards/rejected": -0.8675751090049744, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 8.241758241758242e-07, | |
| "logits/chosen": -4.553553581237793, | |
| "logits/rejected": -4.103464603424072, | |
| "logps/chosen": -392.54351806640625, | |
| "logps/rejected": -390.0396728515625, | |
| "loss": 0.0861, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.4598144590854645, | |
| "rewards/margins": 2.409454345703125, | |
| "rewards/rejected": -1.949639916419983, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 8.791208791208792e-07, | |
| "logits/chosen": -4.560757637023926, | |
| "logits/rejected": -3.3375790119171143, | |
| "logps/chosen": -458.788818359375, | |
| "logps/rejected": -237.71102905273438, | |
| "loss": 0.1327, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.553814709186554, | |
| "rewards/margins": 1.9525666236877441, | |
| "rewards/rejected": -1.3987518548965454, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.340659340659342e-07, | |
| "logits/chosen": -4.524593830108643, | |
| "logits/rejected": -3.837027072906494, | |
| "logps/chosen": -485.5209045410156, | |
| "logps/rejected": -361.74188232421875, | |
| "loss": 0.0366, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.9675140380859375, | |
| "rewards/margins": 3.29044508934021, | |
| "rewards/rejected": -2.3229310512542725, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.890109890109891e-07, | |
| "logits/chosen": -4.552459239959717, | |
| "logits/rejected": -3.4426755905151367, | |
| "logps/chosen": -637.16259765625, | |
| "logps/rejected": -307.08282470703125, | |
| "loss": 0.024, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.276123046875, | |
| "rewards/margins": 3.7193756103515625, | |
| "rewards/rejected": -2.4432525634765625, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.0439560439560442e-06, | |
| "logits/chosen": -4.615756034851074, | |
| "logits/rejected": -3.5999937057495117, | |
| "logps/chosen": -299.9642333984375, | |
| "logps/rejected": -382.33734130859375, | |
| "loss": 0.015, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.375402808189392, | |
| "rewards/margins": 4.191192626953125, | |
| "rewards/rejected": -2.8157899379730225, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.098901098901099e-06, | |
| "logits/chosen": -4.56257963180542, | |
| "logits/rejected": -3.9953930377960205, | |
| "logps/chosen": -304.43365478515625, | |
| "logps/rejected": -345.88836669921875, | |
| "loss": 0.0122, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.693988025188446, | |
| "rewards/margins": 4.401156425476074, | |
| "rewards/rejected": -3.7071685791015625, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.153846153846154e-06, | |
| "logits/chosen": -4.4777092933654785, | |
| "logits/rejected": -3.514190435409546, | |
| "logps/chosen": -350.09747314453125, | |
| "logps/rejected": -357.66943359375, | |
| "loss": 0.0037, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.7525604963302612, | |
| "rewards/margins": 5.601266384124756, | |
| "rewards/rejected": -3.848706007003784, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.2087912087912089e-06, | |
| "logits/chosen": -4.659963607788086, | |
| "logits/rejected": -3.676948070526123, | |
| "logps/chosen": -426.1668701171875, | |
| "logps/rejected": -333.86212158203125, | |
| "loss": 0.0013, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.552722215652466, | |
| "rewards/margins": 6.66888427734375, | |
| "rewards/rejected": -4.116162300109863, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.2637362637362637e-06, | |
| "logits/chosen": -4.635252952575684, | |
| "logits/rejected": -4.635253429412842, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.3186813186813187e-06, | |
| "logits/chosen": -4.54569149017334, | |
| "logits/rejected": -4.585753440856934, | |
| "logps/chosen": -268.20098876953125, | |
| "logps/rejected": -388.1652526855469, | |
| "loss": 0.0254, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.103280782699585, | |
| "rewards/margins": 3.65966796875, | |
| "rewards/rejected": -1.5563873052597046, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.3736263736263738e-06, | |
| "logits/chosen": -4.528458595275879, | |
| "logits/rejected": -3.756037473678589, | |
| "logps/chosen": -320.74591064453125, | |
| "logps/rejected": -383.32171630859375, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.2982544898986816, | |
| "rewards/margins": 8.030874252319336, | |
| "rewards/rejected": -5.7326202392578125, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "logits/chosen": -4.45478630065918, | |
| "logits/rejected": -3.7227022647857666, | |
| "logps/chosen": -421.645263671875, | |
| "logps/rejected": -337.57232666015625, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.300579786300659, | |
| "rewards/margins": 8.780420303344727, | |
| "rewards/rejected": -6.479840278625488, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.4835164835164837e-06, | |
| "logits/chosen": -4.644472122192383, | |
| "logits/rejected": -3.8435280323028564, | |
| "logps/chosen": -344.5352783203125, | |
| "logps/rejected": -387.8581848144531, | |
| "loss": 0.0008, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.017681837081909, | |
| "rewards/margins": 7.160736083984375, | |
| "rewards/rejected": -5.143054485321045, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "logits/chosen": -4.358231544494629, | |
| "logits/rejected": -3.987455129623413, | |
| "logps/chosen": -300.4159851074219, | |
| "logps/rejected": -426.6040344238281, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.130636692047119, | |
| "rewards/margins": 8.095977783203125, | |
| "rewards/rejected": -5.965341091156006, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.5934065934065933e-06, | |
| "logits/chosen": -4.457190990447998, | |
| "logits/rejected": -3.8622899055480957, | |
| "logps/chosen": -428.80010986328125, | |
| "logps/rejected": -282.08941650390625, | |
| "loss": 0.0036, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.919866919517517, | |
| "rewards/margins": 5.6237030029296875, | |
| "rewards/rejected": -3.70383620262146, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.6483516483516484e-06, | |
| "logits/chosen": -4.373894214630127, | |
| "logits/rejected": -3.93933367729187, | |
| "logps/chosen": -282.36163330078125, | |
| "logps/rejected": -356.90789794921875, | |
| "loss": 0.0004, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.413839817047119, | |
| "rewards/margins": 7.732351779937744, | |
| "rewards/rejected": -5.318511962890625, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.7032967032967034e-06, | |
| "logits/chosen": -4.522324085235596, | |
| "logits/rejected": -3.5009684562683105, | |
| "logps/chosen": -299.6717834472656, | |
| "logps/rejected": -359.8671875, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1251494884490967, | |
| "rewards/margins": 8.901113510131836, | |
| "rewards/rejected": -6.775964260101318, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.7582417582417585e-06, | |
| "logits/chosen": -4.407108783721924, | |
| "logits/rejected": -3.9882192611694336, | |
| "logps/chosen": -279.01580810546875, | |
| "logps/rejected": -309.4888916015625, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.941366672515869, | |
| "rewards/margins": 8.681912422180176, | |
| "rewards/rejected": -5.740545749664307, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.8131868131868133e-06, | |
| "logits/chosen": -4.558385848999023, | |
| "logits/rejected": -4.195313453674316, | |
| "logps/chosen": -283.02545166015625, | |
| "logps/rejected": -402.60540771484375, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.7620728015899658, | |
| "rewards/margins": 9.408578872680664, | |
| "rewards/rejected": -7.646505832672119, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.8681318681318684e-06, | |
| "logits/chosen": -4.52475118637085, | |
| "logits/rejected": -4.018232345581055, | |
| "logps/chosen": -304.9980773925781, | |
| "logps/rejected": -374.7655029296875, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.8536040782928467, | |
| "rewards/margins": 9.533895492553711, | |
| "rewards/rejected": -6.680291652679443, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.9230769230769234e-06, | |
| "logits/chosen": -4.477925777435303, | |
| "logits/rejected": -3.7866082191467285, | |
| "logps/chosen": -269.1673278808594, | |
| "logps/rejected": -379.8326416015625, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.563244581222534, | |
| "rewards/margins": 9.162829399108887, | |
| "rewards/rejected": -6.599585056304932, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.9780219780219782e-06, | |
| "logits/chosen": -4.392153263092041, | |
| "logits/rejected": -4.061153411865234, | |
| "logps/chosen": -323.526611328125, | |
| "logps/rejected": -459.18817138671875, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.4470367431640625, | |
| "rewards/margins": 9.656204223632812, | |
| "rewards/rejected": -7.20916748046875, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 2.032967032967033e-06, | |
| "logits/chosen": -4.585270881652832, | |
| "logits/rejected": -3.468134641647339, | |
| "logps/chosen": -314.48748779296875, | |
| "logps/rejected": -404.564453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.6204469203948975, | |
| "rewards/margins": 10.225598335266113, | |
| "rewards/rejected": -7.605151653289795, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.0879120879120883e-06, | |
| "logits/chosen": -4.52483606338501, | |
| "logits/rejected": -3.6769773960113525, | |
| "logps/chosen": -283.4615173339844, | |
| "logps/rejected": -368.316650390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.3265349864959717, | |
| "rewards/margins": 9.931375503540039, | |
| "rewards/rejected": -7.604840278625488, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.1428571428571427e-06, | |
| "logits/chosen": -4.958460330963135, | |
| "logits/rejected": -3.5075302124023438, | |
| "logps/chosen": -1548.0390625, | |
| "logps/rejected": -430.23260498046875, | |
| "loss": 0.0066, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.080407738685608, | |
| "rewards/margins": 5.01118803024292, | |
| "rewards/rejected": -3.9307801723480225, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.197802197802198e-06, | |
| "logits/chosen": -4.469013690948486, | |
| "logits/rejected": -3.7112741470336914, | |
| "logps/chosen": -329.34527587890625, | |
| "logps/rejected": -353.8309326171875, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.11572265625, | |
| "rewards/margins": 8.09234619140625, | |
| "rewards/rejected": -4.97662353515625, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.252747252747253e-06, | |
| "logits/chosen": -4.559028625488281, | |
| "logits/rejected": -3.685147523880005, | |
| "logps/chosen": -350.5050048828125, | |
| "logps/rejected": -432.88720703125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.776409864425659, | |
| "rewards/margins": 11.043191909790039, | |
| "rewards/rejected": -8.2667818069458, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.307692307692308e-06, | |
| "logits/chosen": -4.4963531494140625, | |
| "logits/rejected": -3.1627519130706787, | |
| "logps/chosen": -553.4663696289062, | |
| "logps/rejected": -415.58251953125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.3123841285705566, | |
| "rewards/margins": 13.567768096923828, | |
| "rewards/rejected": -10.255383491516113, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.362637362637363e-06, | |
| "logits/chosen": -4.479090690612793, | |
| "logits/rejected": -3.9051225185394287, | |
| "logps/chosen": -281.47723388671875, | |
| "logps/rejected": -345.96337890625, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1858582496643066, | |
| "rewards/margins": 7.392383098602295, | |
| "rewards/rejected": -5.206524848937988, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.4175824175824177e-06, | |
| "logits/chosen": -4.356232166290283, | |
| "logits/rejected": -3.9175236225128174, | |
| "logps/chosen": -362.4033203125, | |
| "logps/rejected": -447.981201171875, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.4192932844161987, | |
| "rewards/margins": 9.018689155578613, | |
| "rewards/rejected": -7.599395751953125, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.472527472527473e-06, | |
| "logits/chosen": -4.230052471160889, | |
| "logits/rejected": -3.6365554332733154, | |
| "logps/chosen": -336.6274108886719, | |
| "logps/rejected": -369.9189453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9184234142303467, | |
| "rewards/margins": 10.501449584960938, | |
| "rewards/rejected": -7.58302640914917, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.5274725274725274e-06, | |
| "logits/chosen": -4.472602367401123, | |
| "logits/rejected": -3.738085985183716, | |
| "logps/chosen": -372.7972717285156, | |
| "logps/rejected": -331.42022705078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.7339019775390625, | |
| "rewards/margins": 10.597005844116211, | |
| "rewards/rejected": -7.863104343414307, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.582417582417583e-06, | |
| "logits/chosen": -4.685428619384766, | |
| "logits/rejected": -3.7145485877990723, | |
| "logps/chosen": -326.884521484375, | |
| "logps/rejected": -369.75830078125, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.3666748106479645, | |
| "rewards/margins": 8.955459594726562, | |
| "rewards/rejected": -8.588785171508789, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.6373626373626375e-06, | |
| "logits/chosen": -4.467536926269531, | |
| "logits/rejected": -3.8350751399993896, | |
| "logps/chosen": -312.74237060546875, | |
| "logps/rejected": -340.1723937988281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.017587423324585, | |
| "rewards/margins": 10.204119682312012, | |
| "rewards/rejected": -7.186532497406006, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.6923076923076923e-06, | |
| "logits/chosen": -4.286996364593506, | |
| "logits/rejected": -3.920314073562622, | |
| "logps/chosen": -331.720703125, | |
| "logps/rejected": -453.9488830566406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.2590577602386475, | |
| "rewards/margins": 13.077414512634277, | |
| "rewards/rejected": -9.81835651397705, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.7472527472527476e-06, | |
| "logits/chosen": -4.573817729949951, | |
| "logits/rejected": -4.573817729949951, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8021978021978024e-06, | |
| "logits/chosen": -4.560126781463623, | |
| "logits/rejected": -3.358957529067993, | |
| "logps/chosen": -505.0814208984375, | |
| "logps/rejected": -392.2349853515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.12652587890625, | |
| "rewards/margins": 11.479181289672852, | |
| "rewards/rejected": -9.352655410766602, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "logits/chosen": -4.519628047943115, | |
| "logits/rejected": -3.72153902053833, | |
| "logps/chosen": -297.31982421875, | |
| "logps/rejected": -359.1614990234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.254974365234375, | |
| "rewards/margins": 9.919842720031738, | |
| "rewards/rejected": -7.664868354797363, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.9120879120879125e-06, | |
| "logits/chosen": -4.356388568878174, | |
| "logits/rejected": -3.934237241744995, | |
| "logps/chosen": -305.0617370605469, | |
| "logps/rejected": -370.7964172363281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9260528087615967, | |
| "rewards/margins": 10.116022109985352, | |
| "rewards/rejected": -7.189969062805176, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.9670329670329673e-06, | |
| "logits/chosen": -4.416037082672119, | |
| "logits/rejected": -4.145164489746094, | |
| "logps/chosen": -205.00186157226562, | |
| "logps/rejected": -364.52728271484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.6028244495391846, | |
| "rewards/margins": 10.653569221496582, | |
| "rewards/rejected": -8.050745010375977, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 3.021978021978022e-06, | |
| "logits/chosen": -4.462647438049316, | |
| "logits/rejected": -3.762413501739502, | |
| "logps/chosen": -300.01239013671875, | |
| "logps/rejected": -449.92999267578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.1943695545196533, | |
| "rewards/margins": 13.079874038696289, | |
| "rewards/rejected": -9.885504722595215, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "logits/chosen": -4.499337196350098, | |
| "logits/rejected": -3.3283798694610596, | |
| "logps/chosen": -415.0854797363281, | |
| "logps/rejected": -453.6877136230469, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.19559645652771, | |
| "rewards/margins": 13.638141632080078, | |
| "rewards/rejected": -10.442544937133789, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.1318681318681323e-06, | |
| "logits/chosen": -4.473818778991699, | |
| "logits/rejected": -3.3775250911712646, | |
| "logps/chosen": -632.7835083007812, | |
| "logps/rejected": -332.44622802734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.4099366664886475, | |
| "rewards/margins": 10.282444953918457, | |
| "rewards/rejected": -7.872508525848389, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.1868131868131867e-06, | |
| "logits/chosen": -4.390290260314941, | |
| "logits/rejected": -3.326557159423828, | |
| "logps/chosen": -386.25701904296875, | |
| "logps/rejected": -294.870361328125, | |
| "loss": 0.0003, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.3045289516448975, | |
| "rewards/margins": 8.1837739944458, | |
| "rewards/rejected": -5.879245281219482, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.2417582417582424e-06, | |
| "logits/chosen": -4.46834135055542, | |
| "logits/rejected": -4.094131946563721, | |
| "logps/chosen": -334.70654296875, | |
| "logps/rejected": -451.2972412109375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.4573609828948975, | |
| "rewards/margins": 12.338126182556152, | |
| "rewards/rejected": -9.880764961242676, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.2967032967032968e-06, | |
| "logits/chosen": -4.594233989715576, | |
| "logits/rejected": -3.5783090591430664, | |
| "logps/chosen": -278.1133117675781, | |
| "logps/rejected": -505.69195556640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.8014190196990967, | |
| "rewards/margins": 13.821588516235352, | |
| "rewards/rejected": -11.020169258117676, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.3516483516483516e-06, | |
| "logits/chosen": -4.607097625732422, | |
| "logits/rejected": -3.978281021118164, | |
| "logps/chosen": -312.921142578125, | |
| "logps/rejected": -342.41351318359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.903662085533142, | |
| "rewards/margins": 10.762219429016113, | |
| "rewards/rejected": -8.85855770111084, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.406593406593407e-06, | |
| "logits/chosen": -4.4642014503479, | |
| "logits/rejected": -3.5378057956695557, | |
| "logps/chosen": -488.7149658203125, | |
| "logps/rejected": -365.5252380371094, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.4717254638671875, | |
| "rewards/margins": 9.707379341125488, | |
| "rewards/rejected": -8.2356538772583, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.4615384615384617e-06, | |
| "logits/chosen": -4.361936569213867, | |
| "logits/rejected": -4.100497245788574, | |
| "logps/chosen": -287.7290344238281, | |
| "logps/rejected": -451.6803283691406, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.551348924636841, | |
| "rewards/margins": 13.295758247375488, | |
| "rewards/rejected": -10.744409561157227, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.516483516483517e-06, | |
| "logits/chosen": -4.3527302742004395, | |
| "logits/rejected": -3.811634063720703, | |
| "logps/chosen": -360.6308898925781, | |
| "logps/rejected": -448.3175354003906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.832226514816284, | |
| "rewards/margins": 12.640280723571777, | |
| "rewards/rejected": -8.808053970336914, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.5714285714285718e-06, | |
| "logits/chosen": -4.550887584686279, | |
| "logits/rejected": -4.550887584686279, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.6263736263736266e-06, | |
| "logits/chosen": -4.38459587097168, | |
| "logits/rejected": -3.529341697692871, | |
| "logps/chosen": -330.7218017578125, | |
| "logps/rejected": -371.4253234863281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.72430419921875, | |
| "rewards/margins": 10.868155479431152, | |
| "rewards/rejected": -9.143851280212402, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.681318681318682e-06, | |
| "logits/chosen": -4.437158584594727, | |
| "logits/rejected": -4.42078971862793, | |
| "logps/chosen": -243.7357177734375, | |
| "logps/rejected": -423.0533447265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9866089820861816, | |
| "rewards/margins": 12.251775741577148, | |
| "rewards/rejected": -9.265167236328125, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3.7362637362637367e-06, | |
| "logits/chosen": -4.38008451461792, | |
| "logits/rejected": -3.6854677200317383, | |
| "logps/chosen": -255.04901123046875, | |
| "logps/rejected": -424.14111328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.8715972900390625, | |
| "rewards/margins": 14.87622356414795, | |
| "rewards/rejected": -11.004626274108887, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3.7912087912087915e-06, | |
| "logits/chosen": -4.497178077697754, | |
| "logits/rejected": -3.6329329013824463, | |
| "logps/chosen": -349.28289794921875, | |
| "logps/rejected": -381.4493408203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.124981641769409, | |
| "rewards/margins": 10.515064239501953, | |
| "rewards/rejected": -8.390082359313965, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3.846153846153847e-06, | |
| "logits/chosen": -4.524789333343506, | |
| "logits/rejected": -3.4265801906585693, | |
| "logps/chosen": -365.79644775390625, | |
| "logps/rejected": -475.97869873046875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.8129639625549316, | |
| "rewards/margins": 14.072290420532227, | |
| "rewards/rejected": -11.259325981140137, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.901098901098901e-06, | |
| "logits/chosen": -4.497722625732422, | |
| "logits/rejected": -3.3662238121032715, | |
| "logps/chosen": -472.6344909667969, | |
| "logps/rejected": -421.3455505371094, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.5328400135040283, | |
| "rewards/margins": 12.811511039733887, | |
| "rewards/rejected": -9.278671264648438, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.9560439560439565e-06, | |
| "logits/chosen": -4.341458797454834, | |
| "logits/rejected": -3.5533199310302734, | |
| "logps/chosen": -340.4435119628906, | |
| "logps/rejected": -364.519775390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.8503997325897217, | |
| "rewards/margins": 11.756946563720703, | |
| "rewards/rejected": -8.906546592712402, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.010989010989012e-06, | |
| "logits/chosen": -4.497629165649414, | |
| "logits/rejected": -4.373428821563721, | |
| "logps/chosen": -253.01666259765625, | |
| "logps/rejected": -411.61370849609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.3009033203125, | |
| "rewards/margins": 12.612372398376465, | |
| "rewards/rejected": -9.311469078063965, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.065934065934066e-06, | |
| "logits/chosen": -4.67010498046875, | |
| "logits/rejected": -4.221648216247559, | |
| "logps/chosen": -267.68695068359375, | |
| "logps/rejected": -406.31768798828125, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.4193084239959717, | |
| "rewards/margins": 8.557241439819336, | |
| "rewards/rejected": -5.137933254241943, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.120879120879121e-06, | |
| "logits/chosen": -4.276376724243164, | |
| "logits/rejected": -3.5325844287872314, | |
| "logps/chosen": -359.96893310546875, | |
| "logps/rejected": -416.0820617675781, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9439544677734375, | |
| "rewards/margins": 13.037528038024902, | |
| "rewards/rejected": -10.093573570251465, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.175824175824177e-06, | |
| "logits/chosen": -4.5093512535095215, | |
| "logits/rejected": -4.4456963539123535, | |
| "logps/chosen": -192.96502685546875, | |
| "logps/rejected": -362.227783203125, | |
| "loss": 0.0007, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.3808960020542145, | |
| "rewards/margins": 7.196664810180664, | |
| "rewards/rejected": -6.815768718719482, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.230769230769231e-06, | |
| "logits/chosen": -4.7614617347717285, | |
| "logits/rejected": -4.761462211608887, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.2857142857142855e-06, | |
| "logits/chosen": -4.45823860168457, | |
| "logits/rejected": -3.9341721534729004, | |
| "logps/chosen": -418.939697265625, | |
| "logps/rejected": -429.19378662109375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.361285448074341, | |
| "rewards/margins": 14.282876968383789, | |
| "rewards/rejected": -11.921591758728027, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.340659340659341e-06, | |
| "logits/chosen": -4.4410505294799805, | |
| "logits/rejected": -4.185164928436279, | |
| "logps/chosen": -232.5185546875, | |
| "logps/rejected": -447.5303649902344, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.2920806407928467, | |
| "rewards/margins": 13.843243598937988, | |
| "rewards/rejected": -11.551162719726562, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.395604395604396e-06, | |
| "logits/chosen": -4.416217803955078, | |
| "logits/rejected": -3.6590464115142822, | |
| "logps/chosen": -433.36480712890625, | |
| "logps/rejected": -375.2690734863281, | |
| "loss": 0.0006, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.0846405029296875, | |
| "rewards/margins": 7.43996000289917, | |
| "rewards/rejected": -6.355319499969482, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.45054945054945e-06, | |
| "logits/chosen": -4.3372907638549805, | |
| "logits/rejected": -4.462950229644775, | |
| "logps/chosen": -397.9360656738281, | |
| "logps/rejected": -463.2792663574219, | |
| "loss": 0.0016, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.9072632193565369, | |
| "rewards/margins": 6.424984931945801, | |
| "rewards/rejected": -7.332248210906982, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.505494505494506e-06, | |
| "logits/chosen": -4.51932954788208, | |
| "logits/rejected": -3.1768529415130615, | |
| "logps/chosen": -661.5889282226562, | |
| "logps/rejected": -380.08734130859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.3720276355743408, | |
| "rewards/margins": 14.595202445983887, | |
| "rewards/rejected": -13.223175048828125, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.560439560439561e-06, | |
| "logits/chosen": -4.376767158508301, | |
| "logits/rejected": -3.4610347747802734, | |
| "logps/chosen": -433.4891357421875, | |
| "logps/rejected": -476.48748779296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.76708984375, | |
| "rewards/margins": 15.365961074829102, | |
| "rewards/rejected": -13.598871231079102, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.615384615384616e-06, | |
| "logits/chosen": -4.398746490478516, | |
| "logits/rejected": -3.6896204948425293, | |
| "logps/chosen": -230.8444061279297, | |
| "logps/rejected": -394.7328796386719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.5491745471954346, | |
| "rewards/margins": 15.824689865112305, | |
| "rewards/rejected": -13.27551555633545, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.6703296703296706e-06, | |
| "logits/chosen": -4.4249114990234375, | |
| "logits/rejected": -3.499051570892334, | |
| "logps/chosen": -363.8670349121094, | |
| "logps/rejected": -453.15826416015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.3732330799102783, | |
| "rewards/margins": 15.843539237976074, | |
| "rewards/rejected": -13.470306396484375, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.725274725274726e-06, | |
| "logits/chosen": -4.579809665679932, | |
| "logits/rejected": -3.963841199874878, | |
| "logps/chosen": -376.39263916015625, | |
| "logps/rejected": -475.49981689453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.1896270513534546, | |
| "rewards/margins": 17.53704261779785, | |
| "rewards/rejected": -16.347415924072266, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.780219780219781e-06, | |
| "logits/chosen": -4.374417304992676, | |
| "logits/rejected": -4.248178005218506, | |
| "logps/chosen": -194.196533203125, | |
| "logps/rejected": -417.26776123046875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.190634250640869, | |
| "rewards/margins": 15.594205856323242, | |
| "rewards/rejected": -13.403571128845215, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.8351648351648355e-06, | |
| "logits/chosen": -4.6437482833862305, | |
| "logits/rejected": -3.73307728767395, | |
| "logps/chosen": -413.7351379394531, | |
| "logps/rejected": -395.78302001953125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.820913791656494, | |
| "rewards/margins": 16.67730712890625, | |
| "rewards/rejected": -12.856393814086914, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.890109890109891e-06, | |
| "logits/chosen": -4.21191930770874, | |
| "logits/rejected": -3.5639426708221436, | |
| "logps/chosen": -447.6401062011719, | |
| "logps/rejected": -457.892578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.4168548583984375, | |
| "rewards/margins": 14.87121868133545, | |
| "rewards/rejected": -13.454363822937012, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.945054945054946e-06, | |
| "logits/chosen": -4.551489353179932, | |
| "logits/rejected": -4.471848964691162, | |
| "logps/chosen": -240.43063354492188, | |
| "logps/rejected": -433.2778625488281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.8902206420898438, | |
| "rewards/margins": 14.926170349121094, | |
| "rewards/rejected": -14.03594970703125, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 5e-06, | |
| "logits/chosen": -4.397589206695557, | |
| "logits/rejected": -4.0388078689575195, | |
| "logps/chosen": -348.8073425292969, | |
| "logps/rejected": -452.66693115234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.624746799468994, | |
| "rewards/margins": 15.00168228149414, | |
| "rewards/rejected": -11.376935005187988, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.999981426489179e-06, | |
| "logits/chosen": -4.5778608322143555, | |
| "logits/rejected": -3.8527991771698, | |
| "logps/chosen": -359.4041748046875, | |
| "logps/rejected": -409.2432861328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.038177490234375, | |
| "rewards/margins": 16.7247314453125, | |
| "rewards/rejected": -13.686553955078125, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.999925706232695e-06, | |
| "logits/chosen": -4.530111312866211, | |
| "logits/rejected": -4.155897617340088, | |
| "logps/chosen": -234.99557495117188, | |
| "logps/rejected": -426.90765380859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.2188384532928467, | |
| "rewards/margins": 14.525558471679688, | |
| "rewards/rejected": -11.306719779968262, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.9998328400584864e-06, | |
| "logits/chosen": -4.477783679962158, | |
| "logits/rejected": -4.133589267730713, | |
| "logps/chosen": -248.23715209960938, | |
| "logps/rejected": -445.22174072265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.41404128074646, | |
| "rewards/margins": 16.36020851135254, | |
| "rewards/rejected": -12.9461669921875, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.999702829346432e-06, | |
| "logits/chosen": -4.469297885894775, | |
| "logits/rejected": -4.458068370819092, | |
| "logps/chosen": -243.61090087890625, | |
| "logps/rejected": -386.2374267578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.7676605582237244, | |
| "rewards/margins": 11.3004789352417, | |
| "rewards/rejected": -10.532818794250488, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.999535676028338e-06, | |
| "logits/chosen": -4.551413536071777, | |
| "logits/rejected": -3.6586270332336426, | |
| "logps/chosen": -347.52716064453125, | |
| "logps/rejected": -452.22930908203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1838836669921875, | |
| "rewards/margins": 17.132701873779297, | |
| "rewards/rejected": -14.948819160461426, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.999331382587901e-06, | |
| "logits/chosen": -4.324283599853516, | |
| "logits/rejected": -3.1241581439971924, | |
| "logps/chosen": -483.8208923339844, | |
| "logps/rejected": -419.9326171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.6294128894805908, | |
| "rewards/margins": 13.845037460327148, | |
| "rewards/rejected": -12.215624809265137, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.999089952060681e-06, | |
| "logits/chosen": -4.464879035949707, | |
| "logits/rejected": -3.448148488998413, | |
| "logps/chosen": -426.8560485839844, | |
| "logps/rejected": -436.4939270019531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1339569091796875, | |
| "rewards/margins": 15.621515274047852, | |
| "rewards/rejected": -13.487558364868164, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.998811388034046e-06, | |
| "logits/chosen": -4.5417256355285645, | |
| "logits/rejected": -4.166463851928711, | |
| "logps/chosen": -389.265625, | |
| "logps/rejected": -494.1796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.934008836746216, | |
| "rewards/margins": 20.474946975708008, | |
| "rewards/rejected": -17.540937423706055, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.998495694647127e-06, | |
| "logits/chosen": -4.398112773895264, | |
| "logits/rejected": -4.488321781158447, | |
| "logps/chosen": -314.2387390136719, | |
| "logps/rejected": -540.993408203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.8422698974609375, | |
| "rewards/margins": 17.285541534423828, | |
| "rewards/rejected": -15.443270683288574, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.998142876590749e-06, | |
| "logits/chosen": -4.43848180770874, | |
| "logits/rejected": -3.8401870727539062, | |
| "logps/chosen": -357.3565673828125, | |
| "logps/rejected": -495.7008972167969, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.558154284954071, | |
| "rewards/margins": 18.115083694458008, | |
| "rewards/rejected": -17.556928634643555, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.997752939107372e-06, | |
| "logits/chosen": -4.379143714904785, | |
| "logits/rejected": -2.8701581954956055, | |
| "logps/chosen": -384.65484619140625, | |
| "logps/rejected": -369.762939453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.203588843345642, | |
| "rewards/margins": 10.755364418029785, | |
| "rewards/rejected": -9.551775932312012, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.997325887990999e-06, | |
| "logits/chosen": -4.355114936828613, | |
| "logits/rejected": -3.516786575317383, | |
| "logps/chosen": -395.44866943359375, | |
| "logps/rejected": -483.1910095214844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.061553955078125, | |
| "rewards/margins": 16.13475799560547, | |
| "rewards/rejected": -14.073203086853027, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.996861729587103e-06, | |
| "logits/chosen": -4.370433330535889, | |
| "logits/rejected": -3.369537353515625, | |
| "logps/chosen": -388.32037353515625, | |
| "logps/rejected": -497.56427001953125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.486157178878784, | |
| "rewards/margins": 18.97348403930664, | |
| "rewards/rejected": -15.487326622009277, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.996360470792524e-06, | |
| "logits/chosen": -4.528186321258545, | |
| "logits/rejected": -3.8962700366973877, | |
| "logps/chosen": -284.3951110839844, | |
| "logps/rejected": -496.885009765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.5079925060272217, | |
| "rewards/margins": 17.01309585571289, | |
| "rewards/rejected": -13.50510311126709, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.9958221190553705e-06, | |
| "logits/chosen": -4.446348190307617, | |
| "logits/rejected": -4.065671443939209, | |
| "logps/chosen": -374.7364501953125, | |
| "logps/rejected": -565.4706420898438, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.9700074195861816, | |
| "rewards/margins": 15.873197555541992, | |
| "rewards/rejected": -11.903189659118652, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.9952466823749076e-06, | |
| "logits/chosen": -4.366666316986084, | |
| "logits/rejected": -3.5727248191833496, | |
| "logps/chosen": -303.04241943359375, | |
| "logps/rejected": -456.4177551269531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.866119384765625, | |
| "rewards/margins": 15.641772270202637, | |
| "rewards/rejected": -13.775652885437012, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.994634169301439e-06, | |
| "logits/chosen": -4.417354583740234, | |
| "logits/rejected": -4.194676399230957, | |
| "logps/chosen": -246.4547576904297, | |
| "logps/rejected": -482.65960693359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9938981533050537, | |
| "rewards/margins": 16.903947830200195, | |
| "rewards/rejected": -13.910049438476562, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.99398458893618e-06, | |
| "logits/chosen": -4.699535846710205, | |
| "logits/rejected": -4.523982524871826, | |
| "logps/chosen": -260.97991943359375, | |
| "logps/rejected": -504.26416015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.6771606802940369, | |
| "rewards/margins": 14.61356258392334, | |
| "rewards/rejected": -15.290722846984863, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.993297950931121e-06, | |
| "logits/chosen": -4.534379005432129, | |
| "logits/rejected": -3.2884504795074463, | |
| "logps/chosen": -392.5225830078125, | |
| "logps/rejected": -483.8066101074219, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.4821900129318237, | |
| "rewards/margins": 17.772390365600586, | |
| "rewards/rejected": -16.29020118713379, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.992574265488883e-06, | |
| "logits/chosen": -4.4783616065979, | |
| "logits/rejected": -3.9488227367401123, | |
| "logps/chosen": -268.69671630859375, | |
| "logps/rejected": -445.88934326171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.969137668609619, | |
| "rewards/margins": 18.754684448242188, | |
| "rewards/rejected": -14.785547256469727, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.991813543362572e-06, | |
| "logits/chosen": -4.584895610809326, | |
| "logits/rejected": -3.91094970703125, | |
| "logps/chosen": -306.7088623046875, | |
| "logps/rejected": -446.6007995605469, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.993066430091858, | |
| "rewards/margins": 13.759903907775879, | |
| "rewards/rejected": -11.766837120056152, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.991015795855611e-06, | |
| "logits/chosen": -4.340481281280518, | |
| "logits/rejected": -3.854095935821533, | |
| "logps/chosen": -297.0914001464844, | |
| "logps/rejected": -418.73492431640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.8690094351768494, | |
| "rewards/margins": 13.999166488647461, | |
| "rewards/rejected": -13.130157470703125, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.990181034821578e-06, | |
| "logits/chosen": -4.376932144165039, | |
| "logits/rejected": -3.738924264907837, | |
| "logps/chosen": -426.25537109375, | |
| "logps/rejected": -413.3848876953125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.9265228509902954, | |
| "rewards/margins": 14.493753433227539, | |
| "rewards/rejected": -12.567230224609375, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.989309272664026e-06, | |
| "logits/chosen": -4.558651924133301, | |
| "logits/rejected": -3.5922670364379883, | |
| "logps/chosen": -469.27423095703125, | |
| "logps/rejected": -514.3047485351562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.9566650390625, | |
| "rewards/margins": 17.330196380615234, | |
| "rewards/rejected": -14.37353229522705, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.988400522336304e-06, | |
| "logits/chosen": -4.404550075531006, | |
| "logits/rejected": -3.672477960586548, | |
| "logps/chosen": -275.354248046875, | |
| "logps/rejected": -364.86968994140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.3228882551193237, | |
| "rewards/margins": 13.743850708007812, | |
| "rewards/rejected": -12.4209623336792, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.987454797341358e-06, | |
| "logits/chosen": -4.511158466339111, | |
| "logits/rejected": -3.3894896507263184, | |
| "logps/chosen": -320.03155517578125, | |
| "logps/rejected": -423.61590576171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1824281215667725, | |
| "rewards/margins": 14.258829116821289, | |
| "rewards/rejected": -12.076400756835938, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.986472111731536e-06, | |
| "logits/chosen": -4.6026153564453125, | |
| "logits/rejected": -4.254034996032715, | |
| "logps/chosen": -234.12818908691406, | |
| "logps/rejected": -384.76104736328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.7228440046310425, | |
| "rewards/margins": 12.307738304138184, | |
| "rewards/rejected": -10.584894180297852, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.985452480108376e-06, | |
| "logits/chosen": -4.389405727386475, | |
| "logits/rejected": -4.311841011047363, | |
| "logps/chosen": -345.70697021484375, | |
| "logps/rejected": -424.91094970703125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.3869384825229645, | |
| "rewards/margins": 11.403149604797363, | |
| "rewards/rejected": -11.01621150970459, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.984395917622387e-06, | |
| "logits/chosen": -4.487508296966553, | |
| "logits/rejected": -3.7850193977355957, | |
| "logps/chosen": -354.9103698730469, | |
| "logps/rejected": -464.49505615234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.3494720458984375, | |
| "rewards/margins": 15.737408638000488, | |
| "rewards/rejected": -15.38793659210205, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.9833024399728295e-06, | |
| "logits/chosen": -4.406660079956055, | |
| "logits/rejected": -3.810431718826294, | |
| "logps/chosen": -335.26019287109375, | |
| "logps/rejected": -485.70513916015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.31449586153030396, | |
| "rewards/margins": 15.32269287109375, | |
| "rewards/rejected": -15.008196830749512, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.982172063407479e-06, | |
| "logits/chosen": -4.340198993682861, | |
| "logits/rejected": -3.715421199798584, | |
| "logps/chosen": -288.08349609375, | |
| "logps/rejected": -427.89544677734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.296283006668091, | |
| "rewards/margins": 17.811279296875, | |
| "rewards/rejected": -15.514996528625488, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.981004804722384e-06, | |
| "logits/chosen": -4.355514049530029, | |
| "logits/rejected": -4.320943355560303, | |
| "logps/chosen": -344.5791320800781, | |
| "logps/rejected": -434.83441162109375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.7778748273849487, | |
| "rewards/margins": 15.279544830322266, | |
| "rewards/rejected": -13.501669883728027, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.979800681261619e-06, | |
| "logits/chosen": -4.434340476989746, | |
| "logits/rejected": -3.5260894298553467, | |
| "logps/chosen": -304.6591491699219, | |
| "logps/rejected": -368.31622314453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.424444556236267, | |
| "rewards/margins": 12.992469787597656, | |
| "rewards/rejected": -11.568025588989258, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.978559710917024e-06, | |
| "logits/chosen": -4.621590614318848, | |
| "logits/rejected": -4.356657028198242, | |
| "logps/chosen": -253.44842529296875, | |
| "logps/rejected": -314.619140625, | |
| "loss": 0.0065, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.195851132273674, | |
| "rewards/margins": 5.029829502105713, | |
| "rewards/rejected": -4.833978176116943, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.9772819121279395e-06, | |
| "logits/chosen": -4.356565475463867, | |
| "logits/rejected": -4.380559921264648, | |
| "logps/chosen": -248.25204467773438, | |
| "logps/rejected": -454.09014892578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.265228271484375, | |
| "rewards/margins": 15.274236679077148, | |
| "rewards/rejected": -16.539464950561523, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.975967303880933e-06, | |
| "logits/chosen": -4.506335258483887, | |
| "logits/rejected": -3.854314088821411, | |
| "logps/chosen": -328.2158508300781, | |
| "logps/rejected": -472.5008239746094, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.1369415521621704, | |
| "rewards/margins": 17.064838409423828, | |
| "rewards/rejected": -18.201780319213867, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.974615905709518e-06, | |
| "logits/chosen": -4.472858905792236, | |
| "logits/rejected": -3.3297882080078125, | |
| "logps/chosen": -406.04315185546875, | |
| "logps/rejected": -437.1851501464844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.828503429889679, | |
| "rewards/margins": 17.225427627563477, | |
| "rewards/rejected": -16.39692497253418, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.973227737693858e-06, | |
| "logits/chosen": -4.501963138580322, | |
| "logits/rejected": -4.339019298553467, | |
| "logps/chosen": -290.47528076171875, | |
| "logps/rejected": -569.775146484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.762200951576233, | |
| "rewards/margins": 17.014074325561523, | |
| "rewards/rejected": -18.776275634765625, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.971802820460481e-06, | |
| "logits/chosen": -4.366243839263916, | |
| "logits/rejected": -3.672136068344116, | |
| "logps/chosen": -432.7600402832031, | |
| "logps/rejected": -490.27880859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.8038910031318665, | |
| "rewards/margins": 19.19203758239746, | |
| "rewards/rejected": -18.388147354125977, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.970341175181957e-06, | |
| "logits/chosen": -4.175065040588379, | |
| "logits/rejected": -3.344090461730957, | |
| "logps/chosen": -656.5657958984375, | |
| "logps/rejected": -555.3231201171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.54205322265625, | |
| "rewards/margins": 20.33514404296875, | |
| "rewards/rejected": -20.877197265625, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.968842823576592e-06, | |
| "logits/chosen": -4.238167762756348, | |
| "logits/rejected": -3.387505054473877, | |
| "logps/chosen": -430.23291015625, | |
| "logps/rejected": -489.0005798339844, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.47160646319389343, | |
| "rewards/margins": 17.979446411132812, | |
| "rewards/rejected": -18.451053619384766, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.967307787908108e-06, | |
| "logits/chosen": -4.278386116027832, | |
| "logits/rejected": -3.415719509124756, | |
| "logps/chosen": -294.67205810546875, | |
| "logps/rejected": -477.9254150390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.18892823159694672, | |
| "rewards/margins": 17.483989715576172, | |
| "rewards/rejected": -17.67291831970215, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.965736090985305e-06, | |
| "logits/chosen": -4.284654140472412, | |
| "logits/rejected": -3.7332231998443604, | |
| "logps/chosen": -395.2005615234375, | |
| "logps/rejected": -559.507568359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.7245513796806335, | |
| "rewards/margins": 21.60517692565918, | |
| "rewards/rejected": -20.880624771118164, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.964127756161727e-06, | |
| "logits/chosen": -4.413480758666992, | |
| "logits/rejected": -3.4410207271575928, | |
| "logps/chosen": -453.9344177246094, | |
| "logps/rejected": -478.32720947265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.7942901849746704, | |
| "rewards/margins": 19.258975982666016, | |
| "rewards/rejected": -18.464685440063477, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.9624828073353144e-06, | |
| "logits/chosen": -4.612053871154785, | |
| "logits/rejected": -3.6660573482513428, | |
| "logps/chosen": -454.968994140625, | |
| "logps/rejected": -491.28668212890625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.370809942483902, | |
| "rewards/margins": 17.405982971191406, | |
| "rewards/rejected": -17.776792526245117, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.960801268948047e-06, | |
| "logits/chosen": -4.630027770996094, | |
| "logits/rejected": -3.9804086685180664, | |
| "logps/chosen": -334.5860290527344, | |
| "logps/rejected": -470.1821594238281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.2251189947128296, | |
| "rewards/margins": 15.139511108398438, | |
| "rewards/rejected": -13.914392471313477, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.959083165985581e-06, | |
| "logits/chosen": -4.38828706741333, | |
| "logits/rejected": -3.9800713062286377, | |
| "logps/chosen": -251.57284545898438, | |
| "logps/rejected": -336.22308349609375, | |
| "loss": 0.0002, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.5879532098770142, | |
| "rewards/margins": 8.483784675598145, | |
| "rewards/rejected": -9.071738243103027, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.957328523976879e-06, | |
| "logits/chosen": -4.331254482269287, | |
| "logits/rejected": -3.744624376296997, | |
| "logps/chosen": -555.8796997070312, | |
| "logps/rejected": -509.46636962890625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.13983154296875, | |
| "rewards/margins": 23.415903091430664, | |
| "rewards/rejected": -20.276071548461914, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.9555373689938325e-06, | |
| "logits/chosen": -4.3973493576049805, | |
| "logits/rejected": -3.3343820571899414, | |
| "logps/chosen": -488.8701171875, | |
| "logps/rejected": -539.1553344726562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.8417510986328125, | |
| "rewards/margins": 21.32379722595215, | |
| "rewards/rejected": -20.482046127319336, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.95370972765087e-06, | |
| "logits/chosen": -3.7870357036590576, | |
| "logits/rejected": -4.47447395324707, | |
| "logps/chosen": -113.8028793334961, | |
| "logps/rejected": -345.9452819824219, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.3848892152309418, | |
| "rewards/margins": 9.808587074279785, | |
| "rewards/rejected": -9.423697471618652, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.951845627104565e-06, | |
| "logits/chosen": -4.488361835479736, | |
| "logits/rejected": -4.143915176391602, | |
| "logps/chosen": -271.135498046875, | |
| "logps/rejected": -533.8534545898438, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.818115234375, | |
| "rewards/margins": 22.837081909179688, | |
| "rewards/rejected": -21.018966674804688, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.9499450950532305e-06, | |
| "logits/chosen": -4.404188632965088, | |
| "logits/rejected": -3.7939138412475586, | |
| "logps/chosen": -502.817626953125, | |
| "logps/rejected": -596.62646484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.257251024246216, | |
| "rewards/margins": 10.3175048828125, | |
| "rewards/rejected": -13.574755668640137, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.948008159736507e-06, | |
| "logits/chosen": -4.459652423858643, | |
| "logits/rejected": -3.455385208129883, | |
| "logps/chosen": -459.57611083984375, | |
| "logps/rejected": -548.26806640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.31995850801467896, | |
| "rewards/margins": 21.58222198486328, | |
| "rewards/rejected": -21.902179718017578, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.9460348499349485e-06, | |
| "logits/chosen": -4.625619888305664, | |
| "logits/rejected": -3.69688081741333, | |
| "logps/chosen": -304.876953125, | |
| "logps/rejected": -458.91082763671875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.4879150390625, | |
| "rewards/margins": 19.09657859802246, | |
| "rewards/rejected": -19.58449363708496, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.944025194969586e-06, | |
| "logits/chosen": -4.478323459625244, | |
| "logits/rejected": -3.7782227993011475, | |
| "logps/chosen": -332.0870361328125, | |
| "logps/rejected": -599.0357666015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.6869843006134033, | |
| "rewards/margins": 25.880760192871094, | |
| "rewards/rejected": -23.193775177001953, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.941979224701499e-06, | |
| "logits/chosen": -4.613561153411865, | |
| "logits/rejected": -4.613561153411865, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.939896969531367e-06, | |
| "logits/chosen": -4.662393093109131, | |
| "logits/rejected": -4.662393093109131, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.937778460399022e-06, | |
| "logits/chosen": -4.4899797439575195, | |
| "logits/rejected": -3.9571731090545654, | |
| "logps/chosen": -311.52081298828125, | |
| "logps/rejected": -528.713134765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.377178907394409, | |
| "rewards/margins": 24.917617797851562, | |
| "rewards/rejected": -21.54043960571289, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.935623728782986e-06, | |
| "logits/chosen": -4.431125164031982, | |
| "logits/rejected": -3.994685173034668, | |
| "logps/chosen": -318.579833984375, | |
| "logps/rejected": -469.0091552734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.33842775225639343, | |
| "rewards/margins": 20.076812744140625, | |
| "rewards/rejected": -20.415241241455078, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.933432806700004e-06, | |
| "logits/chosen": -4.436576843261719, | |
| "logits/rejected": -3.4724605083465576, | |
| "logps/chosen": -499.44195556640625, | |
| "logps/rejected": -503.2141418457031, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.4404419660568237, | |
| "rewards/margins": 17.919048309326172, | |
| "rewards/rejected": -16.478607177734375, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.931205726704566e-06, | |
| "logits/chosen": -4.575839519500732, | |
| "logits/rejected": -3.600397825241089, | |
| "logps/chosen": -335.92559814453125, | |
| "logps/rejected": -512.950439453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.44965821504592896, | |
| "rewards/margins": 20.810823440551758, | |
| "rewards/rejected": -20.36116600036621, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.928942521888431e-06, | |
| "logits/chosen": -4.4273834228515625, | |
| "logits/rejected": -3.1403005123138428, | |
| "logps/chosen": -641.9811401367188, | |
| "logps/rejected": -565.1820678710938, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.003668213030323386, | |
| "rewards/margins": 20.497615814208984, | |
| "rewards/rejected": -20.493947982788086, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.926643225880123e-06, | |
| "logits/chosen": -4.4851250648498535, | |
| "logits/rejected": -3.762442111968994, | |
| "logps/chosen": -542.037353515625, | |
| "logps/rejected": -583.6822509765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.6874328851699829, | |
| "rewards/margins": 22.295642852783203, | |
| "rewards/rejected": -21.60820960998535, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.924307872844444e-06, | |
| "logits/chosen": -4.5215559005737305, | |
| "logits/rejected": -3.9681222438812256, | |
| "logps/chosen": -452.0367736816406, | |
| "logps/rejected": -570.5191650390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.5621551871299744, | |
| "rewards/margins": 24.197097778320312, | |
| "rewards/rejected": -23.63494300842285, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.921936497481956e-06, | |
| "logits/chosen": -4.452681541442871, | |
| "logits/rejected": -4.026414394378662, | |
| "logps/chosen": -238.1045379638672, | |
| "logps/rejected": -457.61431884765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.171229600906372, | |
| "rewards/margins": 20.798799514770508, | |
| "rewards/rejected": -18.6275691986084, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.919529135028473e-06, | |
| "logits/chosen": -4.6069817543029785, | |
| "logits/rejected": -4.021325588226318, | |
| "logps/chosen": -443.1712951660156, | |
| "logps/rejected": -596.0953979492188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.43758851289749146, | |
| "rewards/margins": 20.0040283203125, | |
| "rewards/rejected": -19.56644058227539, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.917085821254532e-06, | |
| "logits/chosen": -4.430685520172119, | |
| "logits/rejected": -3.408867835998535, | |
| "logps/chosen": -333.0233459472656, | |
| "logps/rejected": -514.176025390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.3141602277755737, | |
| "rewards/margins": 21.7270450592041, | |
| "rewards/rejected": -20.412885665893555, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.914606592464865e-06, | |
| "logits/chosen": -4.614257335662842, | |
| "logits/rejected": -4.173254013061523, | |
| "logps/chosen": -392.7569274902344, | |
| "logps/rejected": -600.0775146484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.3662140369415283, | |
| "rewards/margins": 26.564104080200195, | |
| "rewards/rejected": -25.19788932800293, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.91209148549786e-06, | |
| "logits/chosen": -4.470552444458008, | |
| "logits/rejected": -3.6645894050598145, | |
| "logps/chosen": -278.90374755859375, | |
| "logps/rejected": -585.414306640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.7279815673828125, | |
| "rewards/margins": 22.604557037353516, | |
| "rewards/rejected": -18.876575469970703, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.909540537725007e-06, | |
| "logits/chosen": -4.464867115020752, | |
| "logits/rejected": -3.679206609725952, | |
| "logps/chosen": -399.67230224609375, | |
| "logps/rejected": -459.1192932128906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 3.3072266578674316, | |
| "rewards/margins": 20.884450912475586, | |
| "rewards/rejected": -17.577224731445312, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.906953787050354e-06, | |
| "logits/chosen": -4.506504535675049, | |
| "logits/rejected": -4.483677864074707, | |
| "logps/chosen": -212.6044921875, | |
| "logps/rejected": -577.6050415039062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.012289524078369, | |
| "rewards/margins": 21.269466400146484, | |
| "rewards/rejected": -19.257177352905273, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.904331271909932e-06, | |
| "logits/chosen": -4.631925106048584, | |
| "logits/rejected": -3.576526403427124, | |
| "logps/chosen": -348.4132995605469, | |
| "logps/rejected": -490.9519348144531, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.7574676871299744, | |
| "rewards/margins": 21.938953399658203, | |
| "rewards/rejected": -21.181486129760742, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.901673031271194e-06, | |
| "logits/chosen": -4.463425159454346, | |
| "logits/rejected": -3.718756914138794, | |
| "logps/chosen": -464.51141357421875, | |
| "logps/rejected": -558.5987548828125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.619091808795929, | |
| "rewards/margins": 22.998170852661133, | |
| "rewards/rejected": -22.379079818725586, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.898979104632427e-06, | |
| "logits/chosen": -4.307651996612549, | |
| "logits/rejected": -3.490633010864258, | |
| "logps/chosen": -389.7247314453125, | |
| "logps/rejected": -523.1450805664062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.3679230213165283, | |
| "rewards/margins": 20.981260299682617, | |
| "rewards/rejected": -19.61333656311035, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.8962495320221714e-06, | |
| "logits/chosen": -4.304553031921387, | |
| "logits/rejected": -3.776304006576538, | |
| "logps/chosen": -323.0478515625, | |
| "logps/rejected": -520.6531982421875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.1722412109375, | |
| "rewards/margins": 20.817508697509766, | |
| "rewards/rejected": -18.645267486572266, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.8934843539986266e-06, | |
| "logits/chosen": -4.417074680328369, | |
| "logits/rejected": -3.540210008621216, | |
| "logps/chosen": -322.0855407714844, | |
| "logps/rejected": -536.3630981445312, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.3448638916015625, | |
| "rewards/margins": 22.750934600830078, | |
| "rewards/rejected": -20.406070709228516, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.890683611649041e-06, | |
| "logits/chosen": -4.454306125640869, | |
| "logits/rejected": -3.4941771030426025, | |
| "logps/chosen": -449.34722900390625, | |
| "logps/rejected": -584.3948974609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.965466320514679, | |
| "rewards/margins": 24.43212890625, | |
| "rewards/rejected": -23.466663360595703, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.887847346589111e-06, | |
| "logits/chosen": -4.528407096862793, | |
| "logits/rejected": -4.528407096862793, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.884975600962355e-06, | |
| "logits/chosen": -4.461303234100342, | |
| "logits/rejected": -3.4503564834594727, | |
| "logps/chosen": -350.125, | |
| "logps/rejected": -544.8951416015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.8698486685752869, | |
| "rewards/margins": 22.143985748291016, | |
| "rewards/rejected": -21.274137496948242, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.8820684174394935e-06, | |
| "logits/chosen": -4.2601318359375, | |
| "logits/rejected": -3.9379310607910156, | |
| "logps/chosen": -276.892822265625, | |
| "logps/rejected": -565.9779052734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 4.50067138671875, | |
| "rewards/margins": 27.015552520751953, | |
| "rewards/rejected": -22.514881134033203, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.879125839217808e-06, | |
| "logits/chosen": -4.442978382110596, | |
| "logits/rejected": -4.165872097015381, | |
| "logps/chosen": -276.66558837890625, | |
| "logps/rejected": -485.2083435058594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 2.557089328765869, | |
| "rewards/margins": 19.78055763244629, | |
| "rewards/rejected": -17.223468780517578, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.8761479100205085e-06, | |
| "logits/chosen": -4.185311794281006, | |
| "logits/rejected": -4.506956100463867, | |
| "logps/chosen": -302.510009765625, | |
| "logps/rejected": -368.14508056640625, | |
| "loss": 0.0401, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.349456787109375, | |
| "rewards/margins": 3.197528123855591, | |
| "rewards/rejected": -2.848071336746216, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.873134674096072e-06, | |
| "logits/chosen": -4.432115077972412, | |
| "logits/rejected": -4.062368392944336, | |
| "logps/chosen": -437.7923583984375, | |
| "logps/rejected": -577.2886962890625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.143060207366943, | |
| "rewards/margins": 21.335046768188477, | |
| "rewards/rejected": -26.478107452392578, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.870086176217597e-06, | |
| "logits/chosen": -4.418601989746094, | |
| "logits/rejected": -4.349551677703857, | |
| "logps/chosen": -278.1993713378906, | |
| "logps/rejected": -625.385986328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.728192090988159, | |
| "rewards/margins": 23.60893440246582, | |
| "rewards/rejected": -27.337125778198242, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.867002461682129e-06, | |
| "logits/chosen": -4.347306251525879, | |
| "logits/rejected": -4.1640944480896, | |
| "logps/chosen": -397.3504638671875, | |
| "logps/rejected": -576.0902099609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.7298157215118408, | |
| "rewards/margins": 20.283533096313477, | |
| "rewards/rejected": -22.013349533081055, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.863883576309991e-06, | |
| "logits/chosen": -4.444037437438965, | |
| "logits/rejected": -3.735938310623169, | |
| "logps/chosen": -439.25177001953125, | |
| "logps/rejected": -506.2371826171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.559460401535034, | |
| "rewards/margins": 16.74394416809082, | |
| "rewards/rejected": -20.303403854370117, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.860729566444106e-06, | |
| "logits/chosen": -4.453701972961426, | |
| "logits/rejected": -3.483694076538086, | |
| "logps/chosen": -435.145751953125, | |
| "logps/rejected": -411.5998840332031, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.5900208950042725, | |
| "rewards/margins": 14.451112747192383, | |
| "rewards/rejected": -18.041133880615234, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.857540478949302e-06, | |
| "logits/chosen": -4.345010757446289, | |
| "logits/rejected": -3.3772659301757812, | |
| "logps/chosen": -424.6817626953125, | |
| "logps/rejected": -536.5455322265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.265347480773926, | |
| "rewards/margins": 19.152793884277344, | |
| "rewards/rejected": -23.418142318725586, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.854316361211619e-06, | |
| "logits/chosen": -4.456699848175049, | |
| "logits/rejected": -4.032180309295654, | |
| "logps/chosen": -479.0955505371094, | |
| "logps/rejected": -602.84130859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.73911452293396, | |
| "rewards/margins": 17.996774673461914, | |
| "rewards/rejected": -21.735889434814453, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.851057261137608e-06, | |
| "logits/chosen": -4.340909481048584, | |
| "logits/rejected": -3.292670726776123, | |
| "logps/chosen": -462.56268310546875, | |
| "logps/rejected": -523.4163208007812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.709277391433716, | |
| "rewards/margins": 19.967500686645508, | |
| "rewards/rejected": -22.67677879333496, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.847763227153612e-06, | |
| "logits/chosen": -4.428055286407471, | |
| "logits/rejected": -3.5385830402374268, | |
| "logps/chosen": -422.267822265625, | |
| "logps/rejected": -485.8017578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.4887635707855225, | |
| "rewards/margins": 19.343963623046875, | |
| "rewards/rejected": -21.832727432250977, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.844434308205052e-06, | |
| "logits/chosen": -4.273122310638428, | |
| "logits/rejected": -3.7151196002960205, | |
| "logps/chosen": -546.6180419921875, | |
| "logps/rejected": -715.68896484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.0272216796875, | |
| "rewards/margins": 25.675113677978516, | |
| "rewards/rejected": -30.702335357666016, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.841070553755697e-06, | |
| "logits/chosen": -4.607006072998047, | |
| "logits/rejected": -4.249687671661377, | |
| "logps/chosen": -261.2702941894531, | |
| "logps/rejected": -548.1873779296875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.2804367542266846, | |
| "rewards/margins": 21.665143966674805, | |
| "rewards/rejected": -22.945581436157227, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.837672013786931e-06, | |
| "logits/chosen": -4.519053936004639, | |
| "logits/rejected": -4.209323883056641, | |
| "logps/chosen": -303.30255126953125, | |
| "logps/rejected": -551.4930419921875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.127401828765869, | |
| "rewards/margins": 21.955106735229492, | |
| "rewards/rejected": -25.082508087158203, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.8342387387970105e-06, | |
| "logits/chosen": -4.35404634475708, | |
| "logits/rejected": -3.9764857292175293, | |
| "logps/chosen": -331.4461669921875, | |
| "logps/rejected": -582.4991455078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.6374847888946533, | |
| "rewards/margins": 22.73183250427246, | |
| "rewards/rejected": -25.36931800842285, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.830770779800309e-06, | |
| "logits/chosen": -4.2908101081848145, | |
| "logits/rejected": -3.6881422996520996, | |
| "logps/chosen": -402.2869567871094, | |
| "logps/rejected": -572.225830078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.96708083152771, | |
| "rewards/margins": 22.76718521118164, | |
| "rewards/rejected": -25.73426628112793, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.827268188326567e-06, | |
| "logits/chosen": -4.4686408042907715, | |
| "logits/rejected": -3.9434425830841064, | |
| "logps/chosen": -478.5679626464844, | |
| "logps/rejected": -673.1861572265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.687936305999756, | |
| "rewards/margins": 21.72261619567871, | |
| "rewards/rejected": -26.410552978515625, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.823731016420122e-06, | |
| "logits/chosen": -4.402027606964111, | |
| "logits/rejected": -3.672158718109131, | |
| "logps/chosen": -521.04248046875, | |
| "logps/rejected": -567.2761840820312, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.248925685882568, | |
| "rewards/margins": 16.707509994506836, | |
| "rewards/rejected": -21.956436157226562, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.820159316639133e-06, | |
| "logits/chosen": -4.356218338012695, | |
| "logits/rejected": -4.386761665344238, | |
| "logps/chosen": -312.3507080078125, | |
| "logps/rejected": -663.8659057617188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 1.5384186506271362, | |
| "rewards/margins": 29.62138557434082, | |
| "rewards/rejected": -28.08296775817871, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.816553142054806e-06, | |
| "logits/chosen": -4.566652774810791, | |
| "logits/rejected": -4.032877445220947, | |
| "logps/chosen": -390.8731689453125, | |
| "logps/rejected": -503.9954833984375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.9249114990234375, | |
| "rewards/margins": 17.718976974487305, | |
| "rewards/rejected": -22.643888473510742, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.812912546250595e-06, | |
| "logits/chosen": -4.595418453216553, | |
| "logits/rejected": -3.4654226303100586, | |
| "logps/chosen": -437.74615478515625, | |
| "logps/rejected": -564.276611328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.270282030105591, | |
| "rewards/margins": 22.324874877929688, | |
| "rewards/rejected": -24.595157623291016, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.809237583321421e-06, | |
| "logits/chosen": -4.55130672454834, | |
| "logits/rejected": -3.809107542037964, | |
| "logps/chosen": -288.273681640625, | |
| "logps/rejected": -529.8118286132812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.319549560546875, | |
| "rewards/margins": 21.799253463745117, | |
| "rewards/rejected": -25.118803024291992, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.8055283078728525e-06, | |
| "logits/chosen": -4.273970127105713, | |
| "logits/rejected": -3.532085657119751, | |
| "logps/chosen": -394.9105224609375, | |
| "logps/rejected": -580.9087524414062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.0409668684005737, | |
| "rewards/margins": 24.131925582885742, | |
| "rewards/rejected": -25.17289161682129, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.801784775020303e-06, | |
| "logits/chosen": -4.558983325958252, | |
| "logits/rejected": -3.609147071838379, | |
| "logps/chosen": -425.38165283203125, | |
| "logps/rejected": -579.5765991210938, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.166571140289307, | |
| "rewards/margins": 21.48810577392578, | |
| "rewards/rejected": -25.65467643737793, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.798007040388212e-06, | |
| "logits/chosen": -4.523662567138672, | |
| "logits/rejected": -3.308692693710327, | |
| "logps/chosen": -537.5259399414062, | |
| "logps/rejected": -611.4498901367188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.3234070539474487, | |
| "rewards/margins": 22.504859924316406, | |
| "rewards/rejected": -23.828266143798828, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.794195160109215e-06, | |
| "logits/chosen": -4.554165840148926, | |
| "logits/rejected": -4.412696361541748, | |
| "logps/chosen": -283.873046875, | |
| "logps/rejected": -653.0494384765625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.464404344558716, | |
| "rewards/margins": 24.511882781982422, | |
| "rewards/rejected": -26.976287841796875, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.790349190823313e-06, | |
| "logits/chosen": -4.461995601654053, | |
| "logits/rejected": -4.220631122589111, | |
| "logps/chosen": -378.62567138671875, | |
| "logps/rejected": -555.7577514648438, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -7.84375, | |
| "rewards/margins": 15.135431289672852, | |
| "rewards/rejected": -22.97918128967285, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.786469189677026e-06, | |
| "logits/chosen": -4.626312732696533, | |
| "logits/rejected": -3.6043701171875, | |
| "logps/chosen": -404.515625, | |
| "logps/rejected": -554.8936767578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.731201171875, | |
| "rewards/margins": 19.18717384338379, | |
| "rewards/rejected": -22.91837501525879, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.78255521432255e-06, | |
| "logits/chosen": -4.456379413604736, | |
| "logits/rejected": -4.084070205688477, | |
| "logps/chosen": -233.46734619140625, | |
| "logps/rejected": -546.2238159179688, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.8788391351699829, | |
| "rewards/margins": 24.65411376953125, | |
| "rewards/rejected": -25.5329532623291, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.778607322916896e-06, | |
| "logits/chosen": -4.25200891494751, | |
| "logits/rejected": -4.352073669433594, | |
| "logps/chosen": -287.51422119140625, | |
| "logps/rejected": -514.3617553710938, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.367222785949707, | |
| "rewards/margins": 19.036060333251953, | |
| "rewards/rejected": -24.403284072875977, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.7746255741210256e-06, | |
| "logits/chosen": -4.257689952850342, | |
| "logits/rejected": -3.9616525173187256, | |
| "logps/chosen": -436.54052734375, | |
| "logps/rejected": -558.0086669921875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -6.211642742156982, | |
| "rewards/margins": 16.815370559692383, | |
| "rewards/rejected": -23.027013778686523, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.770610027098983e-06, | |
| "logits/chosen": -4.314263820648193, | |
| "logits/rejected": -3.706144332885742, | |
| "logps/chosen": -385.21514892578125, | |
| "logps/rejected": -575.6632080078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.6782593131065369, | |
| "rewards/margins": 22.845712661743164, | |
| "rewards/rejected": -22.16745376586914, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.766560741517014e-06, | |
| "logits/chosen": -4.261018753051758, | |
| "logits/rejected": -3.49680757522583, | |
| "logps/chosen": -443.4586181640625, | |
| "logps/rejected": -528.631103515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.9242799282073975, | |
| "rewards/margins": 18.376794815063477, | |
| "rewards/rejected": -22.301074981689453, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.762477777542676e-06, | |
| "logits/chosen": -4.405641078948975, | |
| "logits/rejected": -3.945005178451538, | |
| "logps/chosen": -405.1590576171875, | |
| "logps/rejected": -605.96142578125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.02716064453125, | |
| "rewards/margins": 23.706918716430664, | |
| "rewards/rejected": -27.734079360961914, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.7583611958439514e-06, | |
| "logits/chosen": -4.536804676055908, | |
| "logits/rejected": -3.4177980422973633, | |
| "logps/chosen": -509.9901123046875, | |
| "logps/rejected": -620.6497802734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.6092407703399658, | |
| "rewards/margins": 23.246379852294922, | |
| "rewards/rejected": -24.855621337890625, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.75421105758834e-06, | |
| "logits/chosen": -4.660959243774414, | |
| "logits/rejected": -4.660958766937256, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.750027424441949e-06, | |
| "logits/chosen": -4.213469982147217, | |
| "logits/rejected": -4.410857200622559, | |
| "logps/chosen": -280.2644348144531, | |
| "logps/rejected": -588.2990112304688, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.470999151468277, | |
| "rewards/margins": 24.910537719726562, | |
| "rewards/rejected": -24.439538955688477, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.745810358568588e-06, | |
| "logits/chosen": -4.407349586486816, | |
| "logits/rejected": -3.4045121669769287, | |
| "logps/chosen": -651.0145263671875, | |
| "logps/rejected": -581.0377807617188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -9.35540771484375, | |
| "rewards/margins": 13.205488204956055, | |
| "rewards/rejected": -22.560895919799805, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.741559922628828e-06, | |
| "logits/chosen": -4.383808612823486, | |
| "logits/rejected": -3.694657325744629, | |
| "logps/chosen": -627.8776245117188, | |
| "logps/rejected": -550.0055541992188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.867218017578125, | |
| "rewards/margins": 19.274717330932617, | |
| "rewards/rejected": -24.141935348510742, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.7372761797790836e-06, | |
| "logits/chosen": -4.419149875640869, | |
| "logits/rejected": -3.9326839447021484, | |
| "logps/chosen": -456.3083801269531, | |
| "logps/rejected": -572.00732421875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.1375579833984375, | |
| "rewards/margins": 19.8563232421875, | |
| "rewards/rejected": -23.993881225585938, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.732959193670672e-06, | |
| "logits/chosen": -4.390028476715088, | |
| "logits/rejected": -3.866453170776367, | |
| "logps/chosen": -332.904052734375, | |
| "logps/rejected": -565.65673828125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.6805603504180908, | |
| "rewards/margins": 21.7492733001709, | |
| "rewards/rejected": -23.429834365844727, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.728609028448862e-06, | |
| "logits/chosen": -4.234015941619873, | |
| "logits/rejected": -3.2613775730133057, | |
| "logps/chosen": -456.681640625, | |
| "logps/rejected": -508.46441650390625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.307135105133057, | |
| "rewards/margins": 17.87537384033203, | |
| "rewards/rejected": -22.18250846862793, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.7242257487519275e-06, | |
| "logits/chosen": -4.365071773529053, | |
| "logits/rejected": -3.898486852645874, | |
| "logps/chosen": -285.94317626953125, | |
| "logps/rejected": -444.3804931640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.7177032828330994, | |
| "rewards/margins": 18.269119262695312, | |
| "rewards/rejected": -18.9868221282959, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.7198094197101826e-06, | |
| "logits/chosen": -4.433129787445068, | |
| "logits/rejected": -3.400374412536621, | |
| "logps/chosen": -395.3111267089844, | |
| "logps/rejected": -566.3218383789062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.34246826171875, | |
| "rewards/margins": 23.32318687438965, | |
| "rewards/rejected": -25.6656551361084, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.715360106945015e-06, | |
| "logits/chosen": -4.460170269012451, | |
| "logits/rejected": -4.486166477203369, | |
| "logps/chosen": -322.58245849609375, | |
| "logps/rejected": -571.2058715820312, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.377227783203125, | |
| "rewards/margins": 21.1688232421875, | |
| "rewards/rejected": -25.546051025390625, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.710877876567912e-06, | |
| "logits/chosen": -4.677746772766113, | |
| "logits/rejected": -3.628592014312744, | |
| "logps/chosen": -282.5626525878906, | |
| "logps/rejected": -498.66815185546875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.01987624168396, | |
| "rewards/margins": 17.119985580444336, | |
| "rewards/rejected": -19.139862060546875, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.706362795179476e-06, | |
| "logits/chosen": -4.649206638336182, | |
| "logits/rejected": -3.9550185203552246, | |
| "logps/chosen": -317.6220703125, | |
| "logps/rejected": -532.2894897460938, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.211251974105835, | |
| "rewards/margins": 20.70464515686035, | |
| "rewards/rejected": -22.915897369384766, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.70181492986844e-06, | |
| "logits/chosen": -4.574020862579346, | |
| "logits/rejected": -3.7530534267425537, | |
| "logps/chosen": -390.477783203125, | |
| "logps/rejected": -538.0514526367188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.6195312738418579, | |
| "rewards/margins": 23.567447662353516, | |
| "rewards/rejected": -24.186979293823242, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.6972343482106615e-06, | |
| "logits/chosen": -4.381142616271973, | |
| "logits/rejected": -3.9664149284362793, | |
| "logps/chosen": -368.2343444824219, | |
| "logps/rejected": -584.2080688476562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.74643874168396, | |
| "rewards/margins": 23.20481300354004, | |
| "rewards/rejected": -26.951251983642578, | |
| "step": 223 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.6926211182681295e-06, | |
| "logits/chosen": -4.515586853027344, | |
| "logits/rejected": -4.021639823913574, | |
| "logps/chosen": -422.8552551269531, | |
| "logps/rejected": -605.24853515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.3351776599884033, | |
| "rewards/margins": 23.986923217773438, | |
| "rewards/rejected": -27.322101593017578, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.687975308587944e-06, | |
| "logits/chosen": -4.370363235473633, | |
| "logits/rejected": -3.52022385597229, | |
| "logps/chosen": -421.40106201171875, | |
| "logps/rejected": -555.5052490234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.351022243499756, | |
| "rewards/margins": 20.446269989013672, | |
| "rewards/rejected": -25.797292709350586, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.683296988201301e-06, | |
| "logits/chosen": -4.45542049407959, | |
| "logits/rejected": -3.843148946762085, | |
| "logps/chosen": -250.9272003173828, | |
| "logps/rejected": -461.41656494140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.5148300528526306, | |
| "rewards/margins": 22.641338348388672, | |
| "rewards/rejected": -22.126508712768555, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.6785862266224695e-06, | |
| "logits/chosen": -4.4527587890625, | |
| "logits/rejected": -3.8313660621643066, | |
| "logps/chosen": -435.7333068847656, | |
| "logps/rejected": -420.346435546875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.598269939422607, | |
| "rewards/margins": 12.53331184387207, | |
| "rewards/rejected": -18.131582260131836, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.673843093847753e-06, | |
| "logits/chosen": -4.46143913269043, | |
| "logits/rejected": -4.0022873878479, | |
| "logps/chosen": -546.3322143554688, | |
| "logps/rejected": -678.267822265625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -6.684521675109863, | |
| "rewards/margins": 22.149757385253906, | |
| "rewards/rejected": -28.834280014038086, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.669067660354456e-06, | |
| "logits/chosen": -4.453671932220459, | |
| "logits/rejected": -3.783543586730957, | |
| "logps/chosen": -489.1321105957031, | |
| "logps/rejected": -653.801513671875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.456167697906494, | |
| "rewards/margins": 27.995878219604492, | |
| "rewards/rejected": -30.452045440673828, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.664259997099829e-06, | |
| "logits/chosen": -4.382017135620117, | |
| "logits/rejected": -4.235163688659668, | |
| "logps/chosen": -401.96905517578125, | |
| "logps/rejected": -611.9321899414062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.1131255626678467, | |
| "rewards/margins": 23.950090408325195, | |
| "rewards/rejected": -26.063215255737305, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.6594201755200205e-06, | |
| "logits/chosen": -4.401975154876709, | |
| "logits/rejected": -4.047399520874023, | |
| "logps/chosen": -347.8199768066406, | |
| "logps/rejected": -534.321533203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.148288249969482, | |
| "rewards/margins": 15.352182388305664, | |
| "rewards/rejected": -20.500471115112305, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.654548267529015e-06, | |
| "logits/chosen": -4.742393970489502, | |
| "logits/rejected": -3.528773546218872, | |
| "logps/chosen": -461.3231201171875, | |
| "logps/rejected": -502.95330810546875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.834014892578125, | |
| "rewards/margins": 19.609724044799805, | |
| "rewards/rejected": -22.44373893737793, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.64964434551756e-06, | |
| "logits/chosen": -4.29525899887085, | |
| "logits/rejected": -3.517086982727051, | |
| "logps/chosen": -655.9037475585938, | |
| "logps/rejected": -467.7865295410156, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.464410305023193, | |
| "rewards/margins": 16.863277435302734, | |
| "rewards/rejected": -21.327688217163086, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.644708482352093e-06, | |
| "logits/chosen": -4.491517066955566, | |
| "logits/rejected": -3.6783647537231445, | |
| "logps/chosen": -402.5279846191406, | |
| "logps/rejected": -560.4140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.5269012451171875, | |
| "rewards/margins": 17.657962799072266, | |
| "rewards/rejected": -22.184864044189453, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.639740751373663e-06, | |
| "logits/chosen": -4.2944111824035645, | |
| "logits/rejected": -3.600904703140259, | |
| "logps/chosen": -518.21533203125, | |
| "logps/rejected": -547.026611328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.848297119140625, | |
| "rewards/margins": 20.74945068359375, | |
| "rewards/rejected": -24.597747802734375, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.634741226396832e-06, | |
| "logits/chosen": -4.682287693023682, | |
| "logits/rejected": -3.6956443786621094, | |
| "logps/chosen": -314.2425537109375, | |
| "logps/rejected": -630.8653564453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.6687896251678467, | |
| "rewards/margins": 24.917322158813477, | |
| "rewards/rejected": -28.586111068725586, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.629709981708586e-06, | |
| "logits/chosen": -4.4025044441223145, | |
| "logits/rejected": -3.328565835952759, | |
| "logps/chosen": -596.0982666015625, | |
| "logps/rejected": -469.07025146484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -7.942303657531738, | |
| "rewards/margins": 13.943493843078613, | |
| "rewards/rejected": -21.88579750061035, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.624647092067226e-06, | |
| "logits/chosen": -4.347430229187012, | |
| "logits/rejected": -4.287944316864014, | |
| "logps/chosen": -307.51995849609375, | |
| "logps/rejected": -544.3563232421875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.6383118629455566, | |
| "rewards/margins": 22.841541290283203, | |
| "rewards/rejected": -25.4798526763916, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.619552632701263e-06, | |
| "logits/chosen": -4.473455905914307, | |
| "logits/rejected": -4.314274787902832, | |
| "logps/chosen": -277.60906982421875, | |
| "logps/rejected": -504.164306640625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.467631816864014, | |
| "rewards/margins": 18.195838928222656, | |
| "rewards/rejected": -23.663471221923828, | |
| "step": 239 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.614426679308291e-06, | |
| "logits/chosen": -4.148372650146484, | |
| "logits/rejected": -3.744719982147217, | |
| "logps/chosen": -355.4638671875, | |
| "logps/rejected": -454.26422119140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.292315721511841, | |
| "rewards/margins": 16.010753631591797, | |
| "rewards/rejected": -18.303070068359375, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.609269308053872e-06, | |
| "logits/chosen": -4.353799343109131, | |
| "logits/rejected": -3.8214452266693115, | |
| "logps/chosen": -393.32989501953125, | |
| "logps/rejected": -582.2720336914062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.2819581031799316, | |
| "rewards/margins": 22.352136611938477, | |
| "rewards/rejected": -24.63409423828125, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.604080595570399e-06, | |
| "logits/chosen": -4.46610689163208, | |
| "logits/rejected": -3.906358242034912, | |
| "logps/chosen": -322.1328125, | |
| "logps/rejected": -599.1141357421875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.1895081996917725, | |
| "rewards/margins": 23.97051429748535, | |
| "rewards/rejected": -26.160022735595703, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.598860618955957e-06, | |
| "logits/chosen": -4.385786056518555, | |
| "logits/rejected": -4.310035228729248, | |
| "logps/chosen": -400.26947021484375, | |
| "logps/rejected": -559.5322875976562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.774789571762085, | |
| "rewards/margins": 18.254159927368164, | |
| "rewards/rejected": -21.028949737548828, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.5936094557731815e-06, | |
| "logits/chosen": -4.240229606628418, | |
| "logits/rejected": -4.443663597106934, | |
| "logps/chosen": -345.6215515136719, | |
| "logps/rejected": -692.10107421875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.39714357256889343, | |
| "rewards/margins": 27.15097427368164, | |
| "rewards/rejected": -26.753829956054688, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.588327184048099e-06, | |
| "logits/chosen": -4.332690238952637, | |
| "logits/rejected": -3.449024200439453, | |
| "logps/chosen": -496.02508544921875, | |
| "logps/rejected": -615.9970092773438, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.7168214321136475, | |
| "rewards/margins": 22.373443603515625, | |
| "rewards/rejected": -26.09026527404785, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.5830138822689755e-06, | |
| "logits/chosen": -4.444944381713867, | |
| "logits/rejected": -3.7115731239318848, | |
| "logps/chosen": -449.7723083496094, | |
| "logps/rejected": -626.166015625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.5469634532928467, | |
| "rewards/margins": 23.168373107910156, | |
| "rewards/rejected": -26.715335845947266, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.577669629385145e-06, | |
| "logits/chosen": -4.408154487609863, | |
| "logits/rejected": -3.651484489440918, | |
| "logps/chosen": -441.13018798828125, | |
| "logps/rejected": -643.3833618164062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.660876512527466, | |
| "rewards/margins": 22.676677703857422, | |
| "rewards/rejected": -25.337554931640625, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.572294504805841e-06, | |
| "logits/chosen": -4.424633502960205, | |
| "logits/rejected": -3.707329034805298, | |
| "logps/chosen": -487.3986511230469, | |
| "logps/rejected": -554.1790771484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.541256904602051, | |
| "rewards/margins": 15.3828706741333, | |
| "rewards/rejected": -20.92412757873535, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.566888588399007e-06, | |
| "logits/chosen": -4.403136730194092, | |
| "logits/rejected": -3.7492878437042236, | |
| "logps/chosen": -355.76708984375, | |
| "logps/rejected": -554.3052978515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.743695020675659, | |
| "rewards/margins": 20.683734893798828, | |
| "rewards/rejected": -24.42742919921875, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.561451960490123e-06, | |
| "logits/chosen": -4.675843715667725, | |
| "logits/rejected": -4.675843715667725, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.5559847018610034e-06, | |
| "logits/chosen": -4.456602096557617, | |
| "logits/rejected": -3.8642396926879883, | |
| "logps/chosen": -359.24835205078125, | |
| "logps/rejected": -608.4000244140625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.042169213294983, | |
| "rewards/margins": 25.666967391967773, | |
| "rewards/rejected": -26.709136962890625, | |
| "step": 251 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.550486893748596e-06, | |
| "logits/chosen": -4.495731830596924, | |
| "logits/rejected": -4.421840190887451, | |
| "logps/chosen": -300.7923889160156, | |
| "logps/rejected": -369.44671630859375, | |
| "loss": 0.0001, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.9146941900253296, | |
| "rewards/margins": 9.700987815856934, | |
| "rewards/rejected": -11.615681648254395, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.544958617843782e-06, | |
| "logits/chosen": -4.233188629150391, | |
| "logits/rejected": -3.7297332286834717, | |
| "logps/chosen": -473.4989013671875, | |
| "logps/rejected": -492.3461608886719, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -6.732303142547607, | |
| "rewards/margins": 15.883966445922852, | |
| "rewards/rejected": -22.616270065307617, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.539399956290152e-06, | |
| "logits/chosen": -4.557020664215088, | |
| "logits/rejected": -3.960388660430908, | |
| "logps/chosen": -353.52593994140625, | |
| "logps/rejected": -604.6260986328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.712493896484375, | |
| "rewards/margins": 22.881561279296875, | |
| "rewards/rejected": -27.59405517578125, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.533810991682799e-06, | |
| "logits/chosen": -4.580037593841553, | |
| "logits/rejected": -4.580037593841553, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.528191807067074e-06, | |
| "logits/chosen": -4.481337547302246, | |
| "logits/rejected": -4.369873046875, | |
| "logps/chosen": -302.75152587890625, | |
| "logps/rejected": -500.8403625488281, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4375977516174316, | |
| "rewards/margins": 18.154129028320312, | |
| "rewards/rejected": -21.591726303100586, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.522542485937369e-06, | |
| "logits/chosen": -4.541010856628418, | |
| "logits/rejected": -4.120445728302002, | |
| "logps/chosen": -266.5628662109375, | |
| "logps/rejected": -520.7110595703125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.6100586652755737, | |
| "rewards/margins": 22.605920791625977, | |
| "rewards/rejected": -24.215978622436523, | |
| "step": 257 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.516863112235864e-06, | |
| "logits/chosen": -4.432363033294678, | |
| "logits/rejected": -4.100122928619385, | |
| "logps/chosen": -329.5076599121094, | |
| "logps/rejected": -615.78515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.16323547065258026, | |
| "rewards/margins": 28.81184196472168, | |
| "rewards/rejected": -28.97507667541504, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.511153770351288e-06, | |
| "logits/chosen": -4.500782012939453, | |
| "logits/rejected": -3.487731695175171, | |
| "logps/chosen": -464.27020263671875, | |
| "logps/rejected": -605.0491943359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.322766304016113, | |
| "rewards/margins": 23.01436996459961, | |
| "rewards/rejected": -28.33713722229004, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.505414545117658e-06, | |
| "logits/chosen": -4.300138473510742, | |
| "logits/rejected": -3.477233648300171, | |
| "logps/chosen": -533.3462524414062, | |
| "logps/rejected": -552.9212646484375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.31990385055542, | |
| "rewards/margins": 18.3931884765625, | |
| "rewards/rejected": -23.713092803955078, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.499645521813024e-06, | |
| "logits/chosen": -4.700889587402344, | |
| "logits/rejected": -4.700889587402344, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.4938467861582e-06, | |
| "logits/chosen": -4.556081771850586, | |
| "logits/rejected": -4.198794841766357, | |
| "logps/chosen": -371.759033203125, | |
| "logps/rejected": -660.7146606445312, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.564685106277466, | |
| "rewards/margins": 23.74681282043457, | |
| "rewards/rejected": -26.311498641967773, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.488018424315488e-06, | |
| "logits/chosen": -4.297285079956055, | |
| "logits/rejected": -4.350274562835693, | |
| "logps/chosen": -330.6549072265625, | |
| "logps/rejected": -572.1771240234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.0300233364105225, | |
| "rewards/margins": 23.603017807006836, | |
| "rewards/rejected": -26.633041381835938, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.482160522887404e-06, | |
| "logits/chosen": -4.338238716125488, | |
| "logits/rejected": -3.436751127243042, | |
| "logps/chosen": -326.945556640625, | |
| "logps/rejected": -512.1168823242188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.8089507818222046, | |
| "rewards/margins": 21.56464385986328, | |
| "rewards/rejected": -23.373594284057617, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.476273168915382e-06, | |
| "logits/chosen": -5.034451007843018, | |
| "logits/rejected": -3.47391414642334, | |
| "logps/chosen": -1412.2579345703125, | |
| "logps/rejected": -403.28057861328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": 0.26282960176467896, | |
| "rewards/margins": 13.486255645751953, | |
| "rewards/rejected": -13.22342586517334, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.470356449878489e-06, | |
| "logits/chosen": -4.398158073425293, | |
| "logits/rejected": -3.418801784515381, | |
| "logps/chosen": -422.3399658203125, | |
| "logps/rejected": -592.7428588867188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.6789307594299316, | |
| "rewards/margins": 21.556623458862305, | |
| "rewards/rejected": -25.235553741455078, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.464410453692122e-06, | |
| "logits/chosen": -4.643823146820068, | |
| "logits/rejected": -3.279494285583496, | |
| "logps/chosen": -560.9666748046875, | |
| "logps/rejected": -519.3795776367188, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.4105286598205566, | |
| "rewards/margins": 20.03549575805664, | |
| "rewards/rejected": -22.44602394104004, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.458435268706699e-06, | |
| "logits/chosen": -4.174355506896973, | |
| "logits/rejected": -3.4881181716918945, | |
| "logps/chosen": -425.60882568359375, | |
| "logps/rejected": -419.9944763183594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.6146514415740967, | |
| "rewards/margins": 16.718265533447266, | |
| "rewards/rejected": -19.332916259765625, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.452430983706351e-06, | |
| "logits/chosen": -4.392972469329834, | |
| "logits/rejected": -3.702141523361206, | |
| "logps/chosen": -366.05242919921875, | |
| "logps/rejected": -583.062255859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.865545630455017, | |
| "rewards/margins": 23.291095733642578, | |
| "rewards/rejected": -25.156641006469727, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.446397687907601e-06, | |
| "logits/chosen": -4.396094799041748, | |
| "logits/rejected": -4.039773464202881, | |
| "logps/chosen": -341.2452392578125, | |
| "logps/rejected": -622.6715087890625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.08438720554113388, | |
| "rewards/margins": 28.248329162597656, | |
| "rewards/rejected": -28.33271598815918, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.440335470958035e-06, | |
| "logits/chosen": -4.646708011627197, | |
| "logits/rejected": -3.6073200702667236, | |
| "logps/chosen": -276.8403625488281, | |
| "logps/rejected": -427.32745361328125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.796478271484375, | |
| "rewards/margins": 17.417638778686523, | |
| "rewards/rejected": -18.2141170501709, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.434244422934976e-06, | |
| "logits/chosen": -4.129358768463135, | |
| "logits/rejected": -4.367154598236084, | |
| "logps/chosen": -171.61856079101562, | |
| "logps/rejected": -549.8220825195312, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.7979736328125, | |
| "rewards/margins": 24.695449829101562, | |
| "rewards/rejected": -25.493423461914062, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.428124634344141e-06, | |
| "logits/chosen": -4.327607154846191, | |
| "logits/rejected": -3.922853946685791, | |
| "logps/chosen": -338.17413330078125, | |
| "logps/rejected": -567.5420532226562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.001150608062744, | |
| "rewards/margins": 21.21744155883789, | |
| "rewards/rejected": -25.218591690063477, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.421976196118297e-06, | |
| "logits/chosen": -4.414794445037842, | |
| "logits/rejected": -3.8170502185821533, | |
| "logps/chosen": -354.2364501953125, | |
| "logps/rejected": -535.7091064453125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.5273651480674744, | |
| "rewards/margins": 22.748092651367188, | |
| "rewards/rejected": -23.27545738220215, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.415799199615912e-06, | |
| "logits/chosen": -4.355689525604248, | |
| "logits/rejected": -3.721245050430298, | |
| "logps/chosen": -403.55718994140625, | |
| "logps/rejected": -614.30224609375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.988238573074341, | |
| "rewards/margins": 26.3142032623291, | |
| "rewards/rejected": -29.30244255065918, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.409593736619795e-06, | |
| "logits/chosen": -4.30208158493042, | |
| "logits/rejected": -4.249013900756836, | |
| "logps/chosen": -330.4805603027344, | |
| "logps/rejected": -520.9468383789062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.18290114402771, | |
| "rewards/margins": 19.691268920898438, | |
| "rewards/rejected": -21.874170303344727, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.403359899335732e-06, | |
| "logits/chosen": -4.452451705932617, | |
| "logits/rejected": -3.5297634601593018, | |
| "logps/chosen": -407.70440673828125, | |
| "logps/rejected": -531.66455078125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.366229295730591, | |
| "rewards/margins": 19.401885986328125, | |
| "rewards/rejected": -21.768115997314453, | |
| "step": 277 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.39709778039112e-06, | |
| "logits/chosen": -4.624291896820068, | |
| "logits/rejected": -3.607923984527588, | |
| "logps/chosen": -432.9666748046875, | |
| "logps/rejected": -430.0704650878906, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.567608833312988, | |
| "rewards/margins": 15.570273399353027, | |
| "rewards/rejected": -20.137882232666016, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.390807472833585e-06, | |
| "logits/chosen": -4.4363179206848145, | |
| "logits/rejected": -4.285643100738525, | |
| "logps/chosen": -346.4957275390625, | |
| "logps/rejected": -557.8546752929688, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.2618744373321533, | |
| "rewards/margins": 22.152734756469727, | |
| "rewards/rejected": -24.414609909057617, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.384489070129604e-06, | |
| "logits/chosen": -4.686382293701172, | |
| "logits/rejected": -4.411667823791504, | |
| "logps/chosen": -297.61651611328125, | |
| "logps/rejected": -598.6287841796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.362146019935608, | |
| "rewards/margins": 23.59848976135254, | |
| "rewards/rejected": -24.960636138916016, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.378142666163114e-06, | |
| "logits/chosen": -4.398681163787842, | |
| "logits/rejected": -3.3622610569000244, | |
| "logps/chosen": -381.4431457519531, | |
| "logps/rejected": -480.4046630859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.9724457263946533, | |
| "rewards/margins": 19.119897842407227, | |
| "rewards/rejected": -22.092344284057617, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.371768355234116e-06, | |
| "logits/chosen": -4.680546283721924, | |
| "logits/rejected": -4.680546283721924, | |
| "logps/chosen": 0.0, | |
| "logps/rejected": 0.0, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.365366232057279e-06, | |
| "logits/chosen": -4.3958964347839355, | |
| "logits/rejected": -3.414220094680786, | |
| "logps/chosen": -470.550537109375, | |
| "logps/rejected": -456.4300231933594, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.259875535964966, | |
| "rewards/margins": 16.188610076904297, | |
| "rewards/rejected": -19.448486328125, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.358936391760524e-06, | |
| "logits/chosen": -4.416255950927734, | |
| "logits/rejected": -3.4653921127319336, | |
| "logps/chosen": -314.48199462890625, | |
| "logps/rejected": -503.40478515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.4892303943634033, | |
| "rewards/margins": 21.383264541625977, | |
| "rewards/rejected": -23.872495651245117, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.3524789298836175e-06, | |
| "logits/chosen": -4.381935119628906, | |
| "logits/rejected": -3.548987627029419, | |
| "logps/chosen": -456.87030029296875, | |
| "logps/rejected": -514.978515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.680187940597534, | |
| "rewards/margins": 18.483985900878906, | |
| "rewards/rejected": -22.164173126220703, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.345993942376752e-06, | |
| "logits/chosen": -4.319192409515381, | |
| "logits/rejected": -3.6070122718811035, | |
| "logps/chosen": -535.0125122070312, | |
| "logps/rejected": -593.177490234375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.068231105804443, | |
| "rewards/margins": 22.915740966796875, | |
| "rewards/rejected": -26.983972549438477, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.3394815255991135e-06, | |
| "logits/chosen": -4.557577610015869, | |
| "logits/rejected": -4.233284950256348, | |
| "logps/chosen": -340.40478515625, | |
| "logps/rejected": -559.3313598632812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.232684373855591, | |
| "rewards/margins": 15.566218376159668, | |
| "rewards/rejected": -18.79890251159668, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.332941776317458e-06, | |
| "logits/chosen": -4.449789524078369, | |
| "logits/rejected": -3.7085957527160645, | |
| "logps/chosen": -460.09100341796875, | |
| "logps/rejected": -565.1859130859375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4739625453948975, | |
| "rewards/margins": 21.726104736328125, | |
| "rewards/rejected": -25.2000675201416, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.32637479170467e-06, | |
| "logits/chosen": -4.309577941894531, | |
| "logits/rejected": -3.4090349674224854, | |
| "logps/chosen": -488.4316711425781, | |
| "logps/rejected": -556.6356201171875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.466006755828857, | |
| "rewards/margins": 21.675212860107422, | |
| "rewards/rejected": -26.141220092773438, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.319780669338316e-06, | |
| "logits/chosen": -4.309078216552734, | |
| "logits/rejected": -3.30332612991333, | |
| "logps/chosen": -507.0978088378906, | |
| "logps/rejected": -519.831787109375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4428741931915283, | |
| "rewards/margins": 14.903473854064941, | |
| "rewards/rejected": -18.34634780883789, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.313159507199197e-06, | |
| "logits/chosen": -4.414469242095947, | |
| "logits/rejected": -3.480733871459961, | |
| "logps/chosen": -471.8912048339844, | |
| "logps/rejected": -557.726318359375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.649487257003784, | |
| "rewards/margins": 18.816673278808594, | |
| "rewards/rejected": -22.46615982055664, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.306511403669897e-06, | |
| "logits/chosen": -4.4712090492248535, | |
| "logits/rejected": -4.122826099395752, | |
| "logps/chosen": -487.50177001953125, | |
| "logps/rejected": -632.9717407226562, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.310089111328125, | |
| "rewards/margins": 18.895023345947266, | |
| "rewards/rejected": -24.20511245727539, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.299836457533313e-06, | |
| "logits/chosen": -4.493587970733643, | |
| "logits/rejected": -4.127647399902344, | |
| "logps/chosen": -348.9983215332031, | |
| "logps/rejected": -613.3372802734375, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -1.7200958728790283, | |
| "rewards/margins": 21.614328384399414, | |
| "rewards/rejected": -23.33442497253418, | |
| "step": 293 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.293134767971193e-06, | |
| "logits/chosen": -4.367143154144287, | |
| "logits/rejected": -3.770662784576416, | |
| "logps/chosen": -465.8094482421875, | |
| "logps/rejected": -645.4409790039062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.2273621559143066, | |
| "rewards/margins": 24.89282989501953, | |
| "rewards/rejected": -28.12019157409668, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.286406434562659e-06, | |
| "logits/chosen": -4.329665660858154, | |
| "logits/rejected": -4.187549591064453, | |
| "logps/chosen": -327.8598937988281, | |
| "logps/rejected": -573.4243774414062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -0.743408203125, | |
| "rewards/margins": 22.067590713500977, | |
| "rewards/rejected": -22.810998916625977, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.2796515572827305e-06, | |
| "logits/chosen": -4.589941501617432, | |
| "logits/rejected": -3.3189613819122314, | |
| "logps/chosen": -437.3631591796875, | |
| "logps/rejected": -500.00341796875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.332589864730835, | |
| "rewards/margins": 17.248987197875977, | |
| "rewards/rejected": -20.58157730102539, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.2728702365008356e-06, | |
| "logits/chosen": -4.512362957000732, | |
| "logits/rejected": -3.4759039878845215, | |
| "logps/chosen": -452.6970520019531, | |
| "logps/rejected": -566.7788696289062, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.2685546875, | |
| "rewards/margins": 21.10858726501465, | |
| "rewards/rejected": -23.37714195251465, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.266062572979323e-06, | |
| "logits/chosen": -4.406839370727539, | |
| "logits/rejected": -3.6835734844207764, | |
| "logps/chosen": -479.810791015625, | |
| "logps/rejected": -554.9146728515625, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.451641798019409, | |
| "rewards/margins": 18.89285659790039, | |
| "rewards/rejected": -21.344497680664062, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.259228667871963e-06, | |
| "logits/chosen": -4.3280510902404785, | |
| "logits/rejected": -3.3908562660217285, | |
| "logps/chosen": -535.2221069335938, | |
| "logps/rejected": -520.6017456054688, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -2.345489501953125, | |
| "rewards/margins": 21.740737915039062, | |
| "rewards/rejected": -24.086227416992188, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.252368622722443e-06, | |
| "logits/chosen": -4.508797645568848, | |
| "logits/rejected": -4.2981276512146, | |
| "logps/chosen": -387.08135986328125, | |
| "logps/rejected": -548.83203125, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -5.486175537109375, | |
| "rewards/margins": 17.451730728149414, | |
| "rewards/rejected": -22.93790626525879, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.245482539462861e-06, | |
| "logits/chosen": -4.583652973175049, | |
| "logits/rejected": -4.022202491760254, | |
| "logps/chosen": -302.711669921875, | |
| "logps/rejected": -516.1565551757812, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -3.4421050548553467, | |
| "rewards/margins": 20.776533126831055, | |
| "rewards/rejected": -24.218637466430664, | |
| "step": 301 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.2385705204122104e-06, | |
| "logits/chosen": -4.387404441833496, | |
| "logits/rejected": -3.931056261062622, | |
| "logps/chosen": -387.2518310546875, | |
| "logps/rejected": -686.424560546875, | |
| "loss": 0.0, | |
| "rewards/accuracies": 1.0, | |
| "rewards/chosen": -4.06422758102417, | |
| "rewards/margins": 27.97595977783203, | |
| "rewards/rejected": -32.04018783569336, | |
| "step": 302 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 906, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |