| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8063872255489022, | |
| "eval_steps": 500, | |
| "global_step": 404, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001996007984031936, | |
| "grad_norm": 29.364113674215965, | |
| "learning_rate": 1.5686274509803922e-08, | |
| "logits/chosen": -1.6221545934677124, | |
| "logits/rejected": -1.6521912813186646, | |
| "logps/chosen": -0.940773606300354, | |
| "logps/rejected": -0.851032555103302, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00998003992015968, | |
| "grad_norm": 31.405892041917173, | |
| "learning_rate": 7.843137254901961e-08, | |
| "logits/chosen": -1.5107966661453247, | |
| "logits/rejected": -1.3356468677520752, | |
| "logps/chosen": -0.7684795260429382, | |
| "logps/rejected": -0.8418906331062317, | |
| "loss": 0.692, | |
| "rewards/accuracies": 0.515625, | |
| "rewards/chosen": 0.0024527055211365223, | |
| "rewards/margins": 0.004762639291584492, | |
| "rewards/rejected": -0.0023099337704479694, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01996007984031936, | |
| "grad_norm": 32.33090024343072, | |
| "learning_rate": 1.5686274509803921e-07, | |
| "logits/chosen": -1.4846872091293335, | |
| "logits/rejected": -1.3766502141952515, | |
| "logps/chosen": -0.8155986070632935, | |
| "logps/rejected": -0.9123713374137878, | |
| "loss": 0.6918, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.003020394593477249, | |
| "rewards/margins": 0.00013434886932373047, | |
| "rewards/rejected": -0.0031547434628009796, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.029940119760479042, | |
| "grad_norm": 28.485164448197256, | |
| "learning_rate": 2.352941176470588e-07, | |
| "logits/chosen": -1.5661466121673584, | |
| "logits/rejected": -1.494202971458435, | |
| "logps/chosen": -0.9453551173210144, | |
| "logps/rejected": -1.0322425365447998, | |
| "loss": 0.6913, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.0035083256661891937, | |
| "rewards/margins": -0.0018780697137117386, | |
| "rewards/rejected": -0.0016302559524774551, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03992015968063872, | |
| "grad_norm": 38.07958741824287, | |
| "learning_rate": 3.1372549019607843e-07, | |
| "logits/chosen": -1.5159441232681274, | |
| "logits/rejected": -1.506949782371521, | |
| "logps/chosen": -1.148013710975647, | |
| "logps/rejected": -1.2202709913253784, | |
| "loss": 0.6887, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.04793555662035942, | |
| "rewards/margins": 0.017127588391304016, | |
| "rewards/rejected": -0.06506314128637314, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0499001996007984, | |
| "grad_norm": 27.572568095419346, | |
| "learning_rate": 3.92156862745098e-07, | |
| "logits/chosen": -1.4518249034881592, | |
| "logits/rejected": -1.3797534704208374, | |
| "logps/chosen": -0.9855057001113892, | |
| "logps/rejected": -1.0200985670089722, | |
| "loss": 0.691, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.07226995378732681, | |
| "rewards/margins": 0.020743293687701225, | |
| "rewards/rejected": -0.09301324933767319, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.059880239520958084, | |
| "grad_norm": 31.06009539067789, | |
| "learning_rate": 4.705882352941176e-07, | |
| "logits/chosen": -1.499815583229065, | |
| "logits/rejected": -1.5191256999969482, | |
| "logps/chosen": -0.9442195892333984, | |
| "logps/rejected": -1.075582504272461, | |
| "loss": 0.6833, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.08427385240793228, | |
| "rewards/margins": 0.05866638571023941, | |
| "rewards/rejected": -0.14294025301933289, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06986027944111776, | |
| "grad_norm": 30.966967781625176, | |
| "learning_rate": 5.490196078431373e-07, | |
| "logits/chosen": -1.4179936647415161, | |
| "logits/rejected": -1.3696165084838867, | |
| "logps/chosen": -0.8713253140449524, | |
| "logps/rejected": -0.910132884979248, | |
| "loss": 0.6843, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -0.03732147440314293, | |
| "rewards/margins": 0.025124484673142433, | |
| "rewards/rejected": -0.06244595721364021, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.07984031936127745, | |
| "grad_norm": 24.405429771065002, | |
| "learning_rate": 6.274509803921569e-07, | |
| "logits/chosen": -1.500732660293579, | |
| "logits/rejected": -1.4325664043426514, | |
| "logps/chosen": -0.9898351430892944, | |
| "logps/rejected": -1.0182645320892334, | |
| "loss": 0.6711, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": 0.010954104363918304, | |
| "rewards/margins": 0.05013161152601242, | |
| "rewards/rejected": -0.039177510887384415, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08982035928143713, | |
| "grad_norm": 58.76778208087395, | |
| "learning_rate": 7.058823529411765e-07, | |
| "logits/chosen": -1.6222164630889893, | |
| "logits/rejected": -1.4863214492797852, | |
| "logps/chosen": -0.9788524508476257, | |
| "logps/rejected": -1.1273555755615234, | |
| "loss": 0.6701, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": 0.04946264624595642, | |
| "rewards/margins": 0.041922032833099365, | |
| "rewards/rejected": 0.007540611084550619, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0998003992015968, | |
| "grad_norm": 32.39288675133871, | |
| "learning_rate": 7.84313725490196e-07, | |
| "logits/chosen": -1.4457480907440186, | |
| "logits/rejected": -1.4977794885635376, | |
| "logps/chosen": -0.9463974237442017, | |
| "logps/rejected": -1.2008764743804932, | |
| "loss": 0.6473, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -0.026257414370775223, | |
| "rewards/margins": 0.270720511674881, | |
| "rewards/rejected": -0.2969779074192047, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10978043912175649, | |
| "grad_norm": 25.562422423609192, | |
| "learning_rate": 7.998440460161416e-07, | |
| "logits/chosen": -1.6372053623199463, | |
| "logits/rejected": -1.4895769357681274, | |
| "logps/chosen": -0.9624001383781433, | |
| "logps/rejected": -1.001164197921753, | |
| "loss": 0.6452, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -0.22816535830497742, | |
| "rewards/margins": 0.20386116206645966, | |
| "rewards/rejected": -0.4320264756679535, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.11976047904191617, | |
| "grad_norm": 24.2619415054151, | |
| "learning_rate": 7.992106913713086e-07, | |
| "logits/chosen": -1.4938465356826782, | |
| "logits/rejected": -1.5313457250595093, | |
| "logps/chosen": -1.0081576108932495, | |
| "logps/rejected": -1.2644726037979126, | |
| "loss": 0.6479, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -0.19145138561725616, | |
| "rewards/margins": 0.5550881624221802, | |
| "rewards/rejected": -0.7465395331382751, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12974051896207583, | |
| "grad_norm": 53.09152893491566, | |
| "learning_rate": 7.980909599927324e-07, | |
| "logits/chosen": -1.6245317459106445, | |
| "logits/rejected": -1.7302849292755127, | |
| "logps/chosen": -0.9516083002090454, | |
| "logps/rejected": -1.0930800437927246, | |
| "loss": 0.6259, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.033558934926986694, | |
| "rewards/margins": 0.2880553603172302, | |
| "rewards/rejected": -0.3216143250465393, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.13972055888223553, | |
| "grad_norm": 22.54639277386895, | |
| "learning_rate": 7.964862161006166e-07, | |
| "logits/chosen": -1.6116857528686523, | |
| "logits/rejected": -1.5303642749786377, | |
| "logps/chosen": -0.9262536764144897, | |
| "logps/rejected": -1.1940580606460571, | |
| "loss": 0.5986, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": 0.052712906152009964, | |
| "rewards/margins": 0.4166509509086609, | |
| "rewards/rejected": -0.3639380633831024, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1497005988023952, | |
| "grad_norm": 42.93936100609094, | |
| "learning_rate": 7.94398414828202e-07, | |
| "logits/chosen": -1.7100902795791626, | |
| "logits/rejected": -1.6730142831802368, | |
| "logps/chosen": -0.9657354354858398, | |
| "logps/rejected": -0.9953187108039856, | |
| "loss": 0.5873, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -0.0733981654047966, | |
| "rewards/margins": 0.29382631182670593, | |
| "rewards/rejected": -0.36722445487976074, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1596806387225549, | |
| "grad_norm": 33.085627858420565, | |
| "learning_rate": 7.918300998397376e-07, | |
| "logits/chosen": -1.5861061811447144, | |
| "logits/rejected": -1.5445070266723633, | |
| "logps/chosen": -0.9483486413955688, | |
| "logps/rejected": -1.051520824432373, | |
| "loss": 0.5927, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.18298718333244324, | |
| "rewards/margins": 0.16031306982040405, | |
| "rewards/rejected": -0.3433002233505249, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16966067864271456, | |
| "grad_norm": 32.604735894329295, | |
| "learning_rate": 7.887844002314185e-07, | |
| "logits/chosen": -1.456621527671814, | |
| "logits/rejected": -1.550982117652893, | |
| "logps/chosen": -0.8949592709541321, | |
| "logps/rejected": -1.0014525651931763, | |
| "loss": 0.6049, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -0.37570685148239136, | |
| "rewards/margins": 0.4748932421207428, | |
| "rewards/rejected": -0.8506000638008118, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.17964071856287425, | |
| "grad_norm": 50.600905445142125, | |
| "learning_rate": 7.852650267190632e-07, | |
| "logits/chosen": -1.6303730010986328, | |
| "logits/rejected": -1.609903335571289, | |
| "logps/chosen": -1.0100046396255493, | |
| "logps/rejected": -1.0304863452911377, | |
| "loss": 0.5861, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.46551984548568726, | |
| "rewards/margins": 0.5207703709602356, | |
| "rewards/rejected": -0.9862901568412781, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.18962075848303392, | |
| "grad_norm": 46.677854954895324, | |
| "learning_rate": 7.812762671171788e-07, | |
| "logits/chosen": -1.8154528141021729, | |
| "logits/rejected": -1.7442315816879272, | |
| "logps/chosen": -1.245895504951477, | |
| "logps/rejected": -1.381966233253479, | |
| "loss": 0.5881, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -0.5333404541015625, | |
| "rewards/margins": 0.9066336750984192, | |
| "rewards/rejected": -1.439974069595337, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1996007984031936, | |
| "grad_norm": 45.70458018922122, | |
| "learning_rate": 7.768229811149186e-07, | |
| "logits/chosen": -1.6239620447158813, | |
| "logits/rejected": -1.7077280282974243, | |
| "logps/chosen": -1.0647861957550049, | |
| "logps/rejected": -1.128450632095337, | |
| "loss": 0.5655, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": 0.02701367810368538, | |
| "rewards/margins": 0.667129635810852, | |
| "rewards/rejected": -0.6401158571243286, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.20958083832335328, | |
| "grad_norm": 68.02288800224188, | |
| "learning_rate": 7.719105943553005e-07, | |
| "logits/chosen": -1.665117859840393, | |
| "logits/rejected": -1.7087205648422241, | |
| "logps/chosen": -0.872536301612854, | |
| "logps/rejected": -1.0017918348312378, | |
| "loss": 0.5936, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -0.5589737892150879, | |
| "rewards/margins": 0.5613604784011841, | |
| "rewards/rejected": -1.1203341484069824, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.21956087824351297, | |
| "grad_norm": 26.445396621363834, | |
| "learning_rate": 7.665450918248958e-07, | |
| "logits/chosen": -1.7527316808700562, | |
| "logits/rejected": -1.701978087425232, | |
| "logps/chosen": -0.9653258323669434, | |
| "logps/rejected": -1.0634543895721436, | |
| "loss": 0.5066, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -0.627923846244812, | |
| "rewards/margins": 0.6550946235656738, | |
| "rewards/rejected": -1.2830184698104858, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.22954091816367264, | |
| "grad_norm": 30.90025286425145, | |
| "learning_rate": 7.607330105620455e-07, | |
| "logits/chosen": -1.7857640981674194, | |
| "logits/rejected": -1.668609619140625, | |
| "logps/chosen": -1.0989948511123657, | |
| "logps/rejected": -1.1228173971176147, | |
| "loss": 0.5595, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -0.6360177993774414, | |
| "rewards/margins": 0.5694562196731567, | |
| "rewards/rejected": -1.2054741382598877, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.23952095808383234, | |
| "grad_norm": 40.011128656457046, | |
| "learning_rate": 7.544814316924858e-07, | |
| "logits/chosen": -1.7470954656600952, | |
| "logits/rejected": -1.745476484298706, | |
| "logps/chosen": -1.1192394495010376, | |
| "logps/rejected": -1.2689249515533447, | |
| "loss": 0.5494, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -1.0388853549957275, | |
| "rewards/margins": 0.6685038805007935, | |
| "rewards/rejected": -1.7073894739151, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.249500998003992, | |
| "grad_norm": 21.89811631617477, | |
| "learning_rate": 7.477979718020876e-07, | |
| "logits/chosen": -1.5766489505767822, | |
| "logits/rejected": -1.4870827198028564, | |
| "logps/chosen": -1.0617902278900146, | |
| "logps/rejected": -1.2029839754104614, | |
| "loss": 0.4894, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -0.9320518374443054, | |
| "rewards/margins": 0.8520485758781433, | |
| "rewards/rejected": -1.7841002941131592, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.25948103792415167, | |
| "grad_norm": 32.50408891288284, | |
| "learning_rate": 7.40690773657219e-07, | |
| "logits/chosen": -1.6710205078125, | |
| "logits/rejected": -1.6744377613067627, | |
| "logps/chosen": -0.9914701581001282, | |
| "logps/rejected": -1.1012578010559082, | |
| "loss": 0.4908, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/chosen": -0.8213608860969543, | |
| "rewards/margins": 0.6687670946121216, | |
| "rewards/rejected": -1.4901279211044312, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2694610778443114, | |
| "grad_norm": 52.90692812097415, | |
| "learning_rate": 7.331684962840398e-07, | |
| "logits/chosen": -1.5417057275772095, | |
| "logits/rejected": -1.563847303390503, | |
| "logps/chosen": -0.9478904008865356, | |
| "logps/rejected": -1.1853077411651611, | |
| "loss": 0.4906, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -0.5953829884529114, | |
| "rewards/margins": 1.0272996425628662, | |
| "rewards/rejected": -1.6226825714111328, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.27944111776447106, | |
| "grad_norm": 29.247151675001287, | |
| "learning_rate": 7.25240304418811e-07, | |
| "logits/chosen": -1.723270058631897, | |
| "logits/rejected": -1.6206943988800049, | |
| "logps/chosen": -1.0017354488372803, | |
| "logps/rejected": -1.1501209735870361, | |
| "loss": 0.4895, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -0.7104414701461792, | |
| "rewards/margins": 1.0606929063796997, | |
| "rewards/rejected": -1.771134376525879, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2894211576846307, | |
| "grad_norm": 23.202561617382436, | |
| "learning_rate": 7.169158573420763e-07, | |
| "logits/chosen": -1.7494615316390991, | |
| "logits/rejected": -1.741146445274353, | |
| "logps/chosen": -1.208905816078186, | |
| "logps/rejected": -1.3130097389221191, | |
| "loss": 0.4886, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.2182276248931885, | |
| "rewards/margins": 1.148219347000122, | |
| "rewards/rejected": -2.3664469718933105, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.2994011976047904, | |
| "grad_norm": 31.417521603405053, | |
| "learning_rate": 7.082052971103157e-07, | |
| "logits/chosen": -1.614556074142456, | |
| "logits/rejected": -1.5841604471206665, | |
| "logps/chosen": -1.156620979309082, | |
| "logps/rejected": -1.2092293500900269, | |
| "loss": 0.5008, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.330456256866455, | |
| "rewards/margins": 0.7478895783424377, | |
| "rewards/rejected": -2.078345775604248, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3093812375249501, | |
| "grad_norm": 22.72615529430511, | |
| "learning_rate": 6.991192361994127e-07, | |
| "logits/chosen": -1.748228669166565, | |
| "logits/rejected": -1.6922461986541748, | |
| "logps/chosen": -1.078489065170288, | |
| "logps/rejected": -1.3014332056045532, | |
| "loss": 0.4438, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -1.2569700479507446, | |
| "rewards/margins": 1.386672854423523, | |
| "rewards/rejected": -2.6436429023742676, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.3193612774451098, | |
| "grad_norm": 53.34962173479164, | |
| "learning_rate": 6.896687445749871e-07, | |
| "logits/chosen": -1.7363059520721436, | |
| "logits/rejected": -1.7595733404159546, | |
| "logps/chosen": -1.115309476852417, | |
| "logps/rejected": -1.239806890487671, | |
| "loss": 0.4622, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -1.4574905633926392, | |
| "rewards/margins": 1.336918830871582, | |
| "rewards/rejected": -2.7944092750549316, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.32934131736526945, | |
| "grad_norm": 40.83000749193078, | |
| "learning_rate": 6.798653362053462e-07, | |
| "logits/chosen": -1.650973916053772, | |
| "logits/rejected": -1.6539020538330078, | |
| "logps/chosen": -1.1761391162872314, | |
| "logps/rejected": -1.4524037837982178, | |
| "loss": 0.4558, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -1.311657190322876, | |
| "rewards/margins": 1.8300256729125977, | |
| "rewards/rejected": -3.1416831016540527, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.3393213572854291, | |
| "grad_norm": 28.447642135314243, | |
| "learning_rate": 6.697209550334893e-07, | |
| "logits/chosen": -1.7830873727798462, | |
| "logits/rejected": -1.8306477069854736, | |
| "logps/chosen": -1.1625844240188599, | |
| "logps/rejected": -1.4296667575836182, | |
| "loss": 0.4682, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -1.1936951875686646, | |
| "rewards/margins": 1.3315410614013672, | |
| "rewards/rejected": -2.525236129760742, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.34930139720558884, | |
| "grad_norm": 42.529786752880625, | |
| "learning_rate": 6.592479604252524e-07, | |
| "logits/chosen": -1.6931123733520508, | |
| "logits/rejected": -1.6376163959503174, | |
| "logps/chosen": -1.1715461015701294, | |
| "logps/rejected": -1.3761537075042725, | |
| "loss": 0.4751, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -1.1568340063095093, | |
| "rewards/margins": 1.0651991367340088, | |
| "rewards/rejected": -2.2220332622528076, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3592814371257485, | |
| "grad_norm": 49.01740947758452, | |
| "learning_rate": 6.484591121113241e-07, | |
| "logits/chosen": -1.7936656475067139, | |
| "logits/rejected": -1.7206109762191772, | |
| "logps/chosen": -1.047082781791687, | |
| "logps/rejected": -1.2329199314117432, | |
| "loss": 0.4627, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.0403250455856323, | |
| "rewards/margins": 1.458751916885376, | |
| "rewards/rejected": -2.499077081680298, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.36926147704590817, | |
| "grad_norm": 33.35075365946081, | |
| "learning_rate": 6.373675546414806e-07, | |
| "logits/chosen": -1.7752662897109985, | |
| "logits/rejected": -1.8046382665634155, | |
| "logps/chosen": -1.105947494506836, | |
| "logps/rejected": -1.4365394115447998, | |
| "loss": 0.4533, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.559922456741333, | |
| "rewards/margins": 1.3131450414657593, | |
| "rewards/rejected": -2.8730673789978027, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.37924151696606784, | |
| "grad_norm": 33.28309177685594, | |
| "learning_rate": 6.259868013699751e-07, | |
| "logits/chosen": -1.7205069065093994, | |
| "logits/rejected": -1.6720272302627563, | |
| "logps/chosen": -1.1314589977264404, | |
| "logps/rejected": -1.2337143421173096, | |
| "loss": 0.4049, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.9102380275726318, | |
| "rewards/margins": 1.2990434169769287, | |
| "rewards/rejected": -3.2092814445495605, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.38922155688622756, | |
| "grad_norm": 22.448955038966368, | |
| "learning_rate": 6.143307179915986e-07, | |
| "logits/chosen": -1.63016676902771, | |
| "logits/rejected": -1.6702827215194702, | |
| "logps/chosen": -1.1162363290786743, | |
| "logps/rejected": -1.364017128944397, | |
| "loss": 0.3814, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -1.8935344219207764, | |
| "rewards/margins": 1.6944122314453125, | |
| "rewards/rejected": -3.5879464149475098, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.3992015968063872, | |
| "grad_norm": 40.67962108560719, | |
| "learning_rate": 6.024135056484655e-07, | |
| "logits/chosen": -1.7706257104873657, | |
| "logits/rejected": -1.7765319347381592, | |
| "logps/chosen": -1.2081105709075928, | |
| "logps/rejected": -1.5039583444595337, | |
| "loss": 0.4225, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -1.8676178455352783, | |
| "rewards/margins": 2.3755626678466797, | |
| "rewards/rejected": -4.243180274963379, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4091816367265469, | |
| "grad_norm": 22.600562301516938, | |
| "learning_rate": 5.902496836281101e-07, | |
| "logits/chosen": -1.772716760635376, | |
| "logits/rejected": -1.7146966457366943, | |
| "logps/chosen": -1.1653116941452026, | |
| "logps/rejected": -1.2957953214645386, | |
| "loss": 0.353, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.8657941818237305, | |
| "rewards/margins": 1.3124840259552002, | |
| "rewards/rejected": -3.1782779693603516, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.41916167664670656, | |
| "grad_norm": 31.666375625024422, | |
| "learning_rate": 5.778540716739709e-07, | |
| "logits/chosen": -1.7554540634155273, | |
| "logits/rejected": -1.8094005584716797, | |
| "logps/chosen": -1.2131294012069702, | |
| "logps/rejected": -1.54654860496521, | |
| "loss": 0.3751, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -1.522384524345398, | |
| "rewards/margins": 2.483349561691284, | |
| "rewards/rejected": -4.005733966827393, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.4291417165668663, | |
| "grad_norm": 23.231350885180493, | |
| "learning_rate": 5.652417719298168e-07, | |
| "logits/chosen": -1.7632023096084595, | |
| "logits/rejected": -1.6775318384170532, | |
| "logps/chosen": -1.254577875137329, | |
| "logps/rejected": -1.4464714527130127, | |
| "loss": 0.42, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -2.3697409629821777, | |
| "rewards/margins": 1.543129324913025, | |
| "rewards/rejected": -3.912869930267334, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.43912175648702595, | |
| "grad_norm": 75.06138455980248, | |
| "learning_rate": 5.524281505401096e-07, | |
| "logits/chosen": -1.7130699157714844, | |
| "logits/rejected": -1.7215654850006104, | |
| "logps/chosen": -1.1106432676315308, | |
| "logps/rejected": -1.389941692352295, | |
| "loss": 0.4627, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -2.0083611011505127, | |
| "rewards/margins": 1.8821561336517334, | |
| "rewards/rejected": -3.890516996383667, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.4491017964071856, | |
| "grad_norm": 38.6593639633161, | |
| "learning_rate": 5.394288189287261e-07, | |
| "logits/chosen": -1.787519097328186, | |
| "logits/rejected": -1.7786083221435547, | |
| "logps/chosen": -1.1749248504638672, | |
| "logps/rejected": -1.4376263618469238, | |
| "loss": 0.3756, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -2.5106921195983887, | |
| "rewards/margins": 2.196367025375366, | |
| "rewards/rejected": -4.707058906555176, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4590818363273453, | |
| "grad_norm": 78.42260836111647, | |
| "learning_rate": 5.26259614778841e-07, | |
| "logits/chosen": -1.642748475074768, | |
| "logits/rejected": -1.6440712213516235, | |
| "logps/chosen": -1.1253970861434937, | |
| "logps/rejected": -1.2666562795639038, | |
| "loss": 0.4147, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -2.194728374481201, | |
| "rewards/margins": 1.729097604751587, | |
| "rewards/rejected": -3.923825740814209, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.469061876247505, | |
| "grad_norm": 29.199406660864277, | |
| "learning_rate": 5.129365827371506e-07, | |
| "logits/chosen": -1.8895553350448608, | |
| "logits/rejected": -1.8804956674575806, | |
| "logps/chosen": -1.1362788677215576, | |
| "logps/rejected": -1.3920345306396484, | |
| "loss": 0.373, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.8095152378082275, | |
| "rewards/margins": 2.366717576980591, | |
| "rewards/rejected": -4.17623233795166, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.47904191616766467, | |
| "grad_norm": 47.16394571127521, | |
| "learning_rate": 4.99475954865942e-07, | |
| "logits/chosen": -1.7921149730682373, | |
| "logits/rejected": -1.7930948734283447, | |
| "logps/chosen": -1.0978577136993408, | |
| "logps/rejected": -1.2884962558746338, | |
| "loss": 0.4121, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -2.472763776779175, | |
| "rewards/margins": 1.3844527006149292, | |
| "rewards/rejected": -3.8572165966033936, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.48902195608782434, | |
| "grad_norm": 49.94184345559498, | |
| "learning_rate": 4.858941308668253e-07, | |
| "logits/chosen": -1.7998384237289429, | |
| "logits/rejected": -1.7332870960235596, | |
| "logps/chosen": -1.2298054695129395, | |
| "logps/rejected": -1.583311915397644, | |
| "loss": 0.3966, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -2.484738349914551, | |
| "rewards/margins": 2.6141140460968018, | |
| "rewards/rejected": -5.09885311126709, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.499001996007984, | |
| "grad_norm": 24.230906422190216, | |
| "learning_rate": 4.72207658100224e-07, | |
| "logits/chosen": -1.8376038074493408, | |
| "logits/rejected": -1.777984619140625, | |
| "logps/chosen": -1.221793532371521, | |
| "logps/rejected": -1.4775179624557495, | |
| "loss": 0.355, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -2.4652364253997803, | |
| "rewards/margins": 2.177865982055664, | |
| "rewards/rejected": -4.643102169036865, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5089820359281437, | |
| "grad_norm": 34.572626217381774, | |
| "learning_rate": 4.5843321142496465e-07, | |
| "logits/chosen": -1.7895371913909912, | |
| "logits/rejected": -1.7715275287628174, | |
| "logps/chosen": -1.2171119451522827, | |
| "logps/rejected": -1.4471348524093628, | |
| "loss": 0.3765, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -2.290224075317383, | |
| "rewards/margins": 1.525314211845398, | |
| "rewards/rejected": -3.8155384063720703, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.5189620758483033, | |
| "grad_norm": 50.38823987307752, | |
| "learning_rate": 4.445875728825301e-07, | |
| "logits/chosen": -1.759508490562439, | |
| "logits/rejected": -1.7566909790039062, | |
| "logps/chosen": -1.2831090688705444, | |
| "logps/rejected": -1.553682565689087, | |
| "loss": 0.407, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -2.9423370361328125, | |
| "rewards/margins": 2.082186222076416, | |
| "rewards/rejected": -5.024522304534912, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5289421157684631, | |
| "grad_norm": 25.180525143347726, | |
| "learning_rate": 4.3068761125072743e-07, | |
| "logits/chosen": -1.7218453884124756, | |
| "logits/rejected": -1.6865999698638916, | |
| "logps/chosen": -1.2321453094482422, | |
| "logps/rejected": -1.4395345449447632, | |
| "loss": 0.378, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/chosen": -2.701998472213745, | |
| "rewards/margins": 1.778617262840271, | |
| "rewards/rejected": -4.480616092681885, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.5389221556886228, | |
| "grad_norm": 34.94821484127908, | |
| "learning_rate": 4.167502614916798e-07, | |
| "logits/chosen": -1.7490154504776, | |
| "logits/rejected": -1.7060623168945312, | |
| "logps/chosen": -1.1463005542755127, | |
| "logps/rejected": -1.4383161067962646, | |
| "loss": 0.3744, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -2.461625576019287, | |
| "rewards/margins": 1.8262317180633545, | |
| "rewards/rejected": -4.287858009338379, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5489021956087824, | |
| "grad_norm": 28.174483103692058, | |
| "learning_rate": 4.027925041191846e-07, | |
| "logits/chosen": -1.6615409851074219, | |
| "logits/rejected": -1.717336893081665, | |
| "logps/chosen": -1.0789427757263184, | |
| "logps/rejected": -1.291114091873169, | |
| "loss": 0.3802, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -2.360185384750366, | |
| "rewards/margins": 1.7126213312149048, | |
| "rewards/rejected": -4.072806358337402, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5588822355289421, | |
| "grad_norm": 65.52620829663537, | |
| "learning_rate": 3.8883134451057253e-07, | |
| "logits/chosen": -1.728354811668396, | |
| "logits/rejected": -1.7853111028671265, | |
| "logps/chosen": -1.1873013973236084, | |
| "logps/rejected": -1.3926045894622803, | |
| "loss": 0.3777, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -2.1488184928894043, | |
| "rewards/margins": 2.161184787750244, | |
| "rewards/rejected": -4.31000280380249, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5688622754491018, | |
| "grad_norm": 28.41327207057278, | |
| "learning_rate": 3.7488379218827464e-07, | |
| "logits/chosen": -1.8485610485076904, | |
| "logits/rejected": -1.7869895696640015, | |
| "logps/chosen": -1.1387436389923096, | |
| "logps/rejected": -1.4143882989883423, | |
| "loss": 0.3322, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -2.3350417613983154, | |
| "rewards/margins": 2.1998612880706787, | |
| "rewards/rejected": -4.534903049468994, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.5788423153692615, | |
| "grad_norm": 54.15837971453539, | |
| "learning_rate": 3.6096684009634026e-07, | |
| "logits/chosen": -1.9337072372436523, | |
| "logits/rejected": -1.8167577981948853, | |
| "logps/chosen": -1.2900314331054688, | |
| "logps/rejected": -1.5611326694488525, | |
| "loss": 0.3949, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -2.437575101852417, | |
| "rewards/margins": 2.304208278656006, | |
| "rewards/rejected": -4.741783142089844, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5888223552894212, | |
| "grad_norm": 30.875206028913684, | |
| "learning_rate": 3.4709744389715095e-07, | |
| "logits/chosen": -1.7289743423461914, | |
| "logits/rejected": -1.7040287256240845, | |
| "logps/chosen": -1.1819794178009033, | |
| "logps/rejected": -1.3941266536712646, | |
| "loss": 0.3588, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -2.6420161724090576, | |
| "rewards/margins": 1.931423544883728, | |
| "rewards/rejected": -4.573439598083496, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.5988023952095808, | |
| "grad_norm": 23.031518202722424, | |
| "learning_rate": 3.332925013135591e-07, | |
| "logits/chosen": -1.7638660669326782, | |
| "logits/rejected": -1.8151865005493164, | |
| "logps/chosen": -1.1131149530410767, | |
| "logps/rejected": -1.3581509590148926, | |
| "loss": 0.3593, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -2.243455171585083, | |
| "rewards/margins": 1.8504345417022705, | |
| "rewards/rejected": -4.0938897132873535, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6087824351297405, | |
| "grad_norm": 52.883386307510996, | |
| "learning_rate": 3.195688315416142e-07, | |
| "logits/chosen": -1.9481338262557983, | |
| "logits/rejected": -1.877634048461914, | |
| "logps/chosen": -1.3473719358444214, | |
| "logps/rejected": -1.5660619735717773, | |
| "loss": 0.3312, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/chosen": -2.013882637023926, | |
| "rewards/margins": 2.0933196544647217, | |
| "rewards/rejected": -4.107202053070068, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.6187624750499002, | |
| "grad_norm": 40.31001639731489, | |
| "learning_rate": 3.0594315475896397e-07, | |
| "logits/chosen": -1.8702493906021118, | |
| "logits/rejected": -1.8942811489105225, | |
| "logps/chosen": -1.2539622783660889, | |
| "logps/rejected": -1.5844415426254272, | |
| "loss": 0.3233, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -2.855128049850464, | |
| "rewards/margins": 2.254939556121826, | |
| "rewards/rejected": -5.110067367553711, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6287425149700598, | |
| "grad_norm": 34.57734687085236, | |
| "learning_rate": 2.924320717538937e-07, | |
| "logits/chosen": -1.9552714824676514, | |
| "logits/rejected": -1.8983757495880127, | |
| "logps/chosen": -1.235822319984436, | |
| "logps/rejected": -1.557267427444458, | |
| "loss": 0.3362, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -2.798520803451538, | |
| "rewards/margins": 2.358416795730591, | |
| "rewards/rejected": -5.156937599182129, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.6387225548902196, | |
| "grad_norm": 22.131955927530566, | |
| "learning_rate": 2.790520436998222e-07, | |
| "logits/chosen": -1.8081642389297485, | |
| "logits/rejected": -1.7835298776626587, | |
| "logps/chosen": -1.2587134838104248, | |
| "logps/rejected": -1.5615605115890503, | |
| "loss": 0.3362, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -2.6179587841033936, | |
| "rewards/margins": 2.341552972793579, | |
| "rewards/rejected": -4.959511756896973, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6487025948103793, | |
| "grad_norm": 25.656535146073534, | |
| "learning_rate": 2.6581937209989804e-07, | |
| "logits/chosen": -1.857273817062378, | |
| "logits/rejected": -1.8463045358657837, | |
| "logps/chosen": -1.207121729850769, | |
| "logps/rejected": -1.4980764389038086, | |
| "loss": 0.2881, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -2.7167863845825195, | |
| "rewards/margins": 2.327130079269409, | |
| "rewards/rejected": -5.043917179107666, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.6586826347305389, | |
| "grad_norm": 80.80388415289183, | |
| "learning_rate": 2.527501789261288e-07, | |
| "logits/chosen": -1.7654087543487549, | |
| "logits/rejected": -1.7988303899765015, | |
| "logps/chosen": -1.2382586002349854, | |
| "logps/rejected": -1.5011346340179443, | |
| "loss": 0.3562, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -2.9657859802246094, | |
| "rewards/margins": 1.982195258140564, | |
| "rewards/rejected": -4.947981357574463, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6686626746506986, | |
| "grad_norm": 58.701745588033376, | |
| "learning_rate": 2.398603869772421e-07, | |
| "logits/chosen": -1.8312063217163086, | |
| "logits/rejected": -1.906640648841858, | |
| "logps/chosen": -1.2148807048797607, | |
| "logps/rejected": -1.609410047531128, | |
| "loss": 0.3771, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -2.625844955444336, | |
| "rewards/margins": 2.595080614089966, | |
| "rewards/rejected": -5.220925807952881, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.6786427145708582, | |
| "grad_norm": 53.98596461549184, | |
| "learning_rate": 2.27165700479207e-07, | |
| "logits/chosen": -1.7870060205459595, | |
| "logits/rejected": -1.8084951639175415, | |
| "logps/chosen": -1.24245023727417, | |
| "logps/rejected": -1.5464107990264893, | |
| "loss": 0.3441, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -3.0009231567382812, | |
| "rewards/margins": 2.439561605453491, | |
| "rewards/rejected": -5.440484523773193, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.688622754491018, | |
| "grad_norm": 31.40057350192729, | |
| "learning_rate": 2.146815859520554e-07, | |
| "logits/chosen": -1.655821442604065, | |
| "logits/rejected": -1.7065486907958984, | |
| "logps/chosen": -1.1306984424591064, | |
| "logps/rejected": -1.4800045490264893, | |
| "loss": 0.3349, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -2.8754384517669678, | |
| "rewards/margins": 2.686849355697632, | |
| "rewards/rejected": -5.562288284301758, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.6986027944111777, | |
| "grad_norm": 31.89392999133678, | |
| "learning_rate": 2.0242325336630769e-07, | |
| "logits/chosen": -1.770398497581482, | |
| "logits/rejected": -1.7964344024658203, | |
| "logps/chosen": -1.3358914852142334, | |
| "logps/rejected": -1.713321328163147, | |
| "loss": 0.3356, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -3.2399375438690186, | |
| "rewards/margins": 2.4227969646453857, | |
| "rewards/rejected": -5.662735462188721, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7085828343313373, | |
| "grad_norm": 36.33203288290445, | |
| "learning_rate": 1.9040563761196839e-07, | |
| "logits/chosen": -1.8168394565582275, | |
| "logits/rejected": -1.877682089805603, | |
| "logps/chosen": -1.2325422763824463, | |
| "logps/rejected": -1.7707788944244385, | |
| "loss": 0.3245, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -2.476655960083008, | |
| "rewards/margins": 3.2025744915008545, | |
| "rewards/rejected": -5.6792311668396, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.718562874251497, | |
| "grad_norm": 43.416666324732795, | |
| "learning_rate": 1.786433803026624e-07, | |
| "logits/chosen": -1.8330310583114624, | |
| "logits/rejected": -1.8215529918670654, | |
| "logps/chosen": -1.3045927286148071, | |
| "logps/rejected": -1.523819088935852, | |
| "loss": 0.3075, | |
| "rewards/accuracies": 0.8374999761581421, | |
| "rewards/chosen": -3.1603009700775146, | |
| "rewards/margins": 1.9945179224014282, | |
| "rewards/rejected": -5.154818534851074, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7285429141716567, | |
| "grad_norm": 41.308059452990165, | |
| "learning_rate": 1.6715081193708426e-07, | |
| "logits/chosen": -1.8217226266860962, | |
| "logits/rejected": -1.7349847555160522, | |
| "logps/chosen": -1.3033679723739624, | |
| "logps/rejected": -1.5060852766036987, | |
| "loss": 0.3159, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -3.0869052410125732, | |
| "rewards/margins": 2.2355074882507324, | |
| "rewards/rejected": -5.322413444519043, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.7385229540918163, | |
| "grad_norm": 42.21028449654826, | |
| "learning_rate": 1.5594193443949295e-07, | |
| "logits/chosen": -1.8006670475006104, | |
| "logits/rejected": -1.7459598779678345, | |
| "logps/chosen": -1.3234542608261108, | |
| "logps/rejected": -1.5523171424865723, | |
| "loss": 0.3351, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -3.036769390106201, | |
| "rewards/margins": 2.387256145477295, | |
| "rewards/rejected": -5.424025535583496, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.7485029940119761, | |
| "grad_norm": 19.788273374335468, | |
| "learning_rate": 1.450304041005241e-07, | |
| "logits/chosen": -1.7887513637542725, | |
| "logits/rejected": -1.7895513772964478, | |
| "logps/chosen": -1.2865995168685913, | |
| "logps/rejected": -1.5641953945159912, | |
| "loss": 0.3256, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -2.724120616912842, | |
| "rewards/margins": 2.6693472862243652, | |
| "rewards/rejected": -5.393467903137207, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.7584830339321357, | |
| "grad_norm": 22.398433956551006, | |
| "learning_rate": 1.3442951493910334e-07, | |
| "logits/chosen": -1.9577217102050781, | |
| "logits/rejected": -1.888419508934021, | |
| "logps/chosen": -1.1813005208969116, | |
| "logps/rejected": -1.487975835800171, | |
| "loss": 0.3656, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -1.874016523361206, | |
| "rewards/margins": 2.6297459602355957, | |
| "rewards/rejected": -4.503762722015381, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.7684630738522954, | |
| "grad_norm": 25.43555535966853, | |
| "learning_rate": 1.2415218250573207e-07, | |
| "logits/chosen": -1.8119207620620728, | |
| "logits/rejected": -1.808342695236206, | |
| "logps/chosen": -1.3371237516403198, | |
| "logps/rejected": -1.5443830490112305, | |
| "loss": 0.3467, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -2.7301838397979736, | |
| "rewards/margins": 2.1141746044158936, | |
| "rewards/rejected": -4.844358921051025, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.7784431137724551, | |
| "grad_norm": 42.565390920889676, | |
| "learning_rate": 1.1421092814687874e-07, | |
| "logits/chosen": -1.92118239402771, | |
| "logits/rejected": -1.880392074584961, | |
| "logps/chosen": -1.2233985662460327, | |
| "logps/rejected": -1.5754501819610596, | |
| "loss": 0.3129, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -2.677239418029785, | |
| "rewards/margins": 2.564099073410034, | |
| "rewards/rejected": -5.24133825302124, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7884231536926147, | |
| "grad_norm": 35.381985967589934, | |
| "learning_rate": 1.0461786374964643e-07, | |
| "logits/chosen": -2.013378620147705, | |
| "logits/rejected": -1.9751570224761963, | |
| "logps/chosen": -1.449527382850647, | |
| "logps/rejected": -1.7529798746109009, | |
| "loss": 0.2973, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -3.0077102184295654, | |
| "rewards/margins": 2.959484577178955, | |
| "rewards/rejected": -5.9671950340271, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.7984031936127745, | |
| "grad_norm": 28.566925881122785, | |
| "learning_rate": 9.538467698530536e-08, | |
| "logits/chosen": -1.760108232498169, | |
| "logits/rejected": -1.8218498229980469, | |
| "logps/chosen": -1.2844475507736206, | |
| "logps/rejected": -1.5693128108978271, | |
| "loss": 0.299, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -2.605151414871216, | |
| "rewards/margins": 2.5104212760925293, | |
| "rewards/rejected": -5.115572452545166, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 501, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 101, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |