| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.47426924660354053, | |
| "eval_steps": 500, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013174145738987238, | |
| "grad_norm": 0.48070791363716125, | |
| "learning_rate": 6.25e-08, | |
| "logits/chosen": 10.32492733001709, | |
| "logits/rejected": 10.282785415649414, | |
| "logps/chosen": -176.09544372558594, | |
| "logps/ref_chosen": -176.09544372558594, | |
| "logps/ref_rejected": -181.75552368164062, | |
| "logps/rejected": -181.75552368164062, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.026348291477974475, | |
| "grad_norm": 0.4185059070587158, | |
| "learning_rate": 1.25e-07, | |
| "logits/chosen": 10.530074119567871, | |
| "logits/rejected": 10.672085762023926, | |
| "logps/chosen": -173.2084503173828, | |
| "logps/ref_chosen": -173.2084503173828, | |
| "logps/ref_rejected": -187.02206420898438, | |
| "logps/rejected": -187.02206420898438, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03952243721696171, | |
| "grad_norm": 0.3848375082015991, | |
| "learning_rate": 1.875e-07, | |
| "logits/chosen": 10.340475082397461, | |
| "logits/rejected": 10.486526489257812, | |
| "logps/chosen": -169.6623992919922, | |
| "logps/ref_chosen": -169.080810546875, | |
| "logps/ref_rejected": -182.52792358398438, | |
| "logps/rejected": -182.80642700195312, | |
| "loss": 0.6947, | |
| "rewards/accuracies": 0.3828125, | |
| "rewards/chosen": -0.005815833806991577, | |
| "rewards/margins": -0.0030306153930723667, | |
| "rewards/rejected": -0.002785218646749854, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05269658295594895, | |
| "grad_norm": 0.7261964678764343, | |
| "learning_rate": 2.5e-07, | |
| "logits/chosen": 10.3652925491333, | |
| "logits/rejected": 10.412069320678711, | |
| "logps/chosen": -177.03428649902344, | |
| "logps/ref_chosen": -176.78369140625, | |
| "logps/ref_rejected": -180.0931396484375, | |
| "logps/rejected": -180.21463012695312, | |
| "loss": 0.6938, | |
| "rewards/accuracies": 0.4765625, | |
| "rewards/chosen": -0.0025058696046471596, | |
| "rewards/margins": -0.0012909012148156762, | |
| "rewards/rejected": -0.0012149682734161615, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06587072869493618, | |
| "grad_norm": 0.3572319746017456, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "logits/chosen": 10.757452011108398, | |
| "logits/rejected": 10.791389465332031, | |
| "logps/chosen": -169.12631225585938, | |
| "logps/ref_chosen": -168.522216796875, | |
| "logps/ref_rejected": -176.08982849121094, | |
| "logps/rejected": -176.59901428222656, | |
| "loss": 0.6936, | |
| "rewards/accuracies": 0.4609375, | |
| "rewards/chosen": -0.0060411859303712845, | |
| "rewards/margins": -0.0009493756806477904, | |
| "rewards/rejected": -0.0050918105989694595, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07904487443392343, | |
| "grad_norm": 0.4071093201637268, | |
| "learning_rate": 3.75e-07, | |
| "logits/chosen": 11.003137588500977, | |
| "logits/rejected": 11.056475639343262, | |
| "logps/chosen": -174.662353515625, | |
| "logps/ref_chosen": -174.5843048095703, | |
| "logps/ref_rejected": -181.65040588378906, | |
| "logps/rejected": -181.72401428222656, | |
| "loss": 0.6932, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.0007804610067978501, | |
| "rewards/margins": -4.43047538283281e-05, | |
| "rewards/rejected": -0.0007361561874859035, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09221902017291066, | |
| "grad_norm": 0.7951124906539917, | |
| "learning_rate": 4.375e-07, | |
| "logits/chosen": 10.003807067871094, | |
| "logits/rejected": 10.067156791687012, | |
| "logps/chosen": -171.4049530029297, | |
| "logps/ref_chosen": -170.4031219482422, | |
| "logps/ref_rejected": -181.00917053222656, | |
| "logps/rejected": -181.8280792236328, | |
| "loss": 0.6941, | |
| "rewards/accuracies": 0.4453125, | |
| "rewards/chosen": -0.010018287226557732, | |
| "rewards/margins": -0.0018291514134034514, | |
| "rewards/rejected": -0.008189136162400246, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1053931659118979, | |
| "grad_norm": 0.5591660737991333, | |
| "learning_rate": 5e-07, | |
| "logits/chosen": 10.874868392944336, | |
| "logits/rejected": 10.79828929901123, | |
| "logps/chosen": -179.36700439453125, | |
| "logps/ref_chosen": -178.47369384765625, | |
| "logps/ref_rejected": -184.52243041992188, | |
| "logps/rejected": -185.44284057617188, | |
| "loss": 0.693, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.008933190256357193, | |
| "rewards/margins": 0.0002710025873966515, | |
| "rewards/rejected": -0.009204192087054253, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.11856731165088513, | |
| "grad_norm": 0.45744186639785767, | |
| "learning_rate": 4.997252228714278e-07, | |
| "logits/chosen": 10.472944259643555, | |
| "logits/rejected": 10.640350341796875, | |
| "logps/chosen": -175.36553955078125, | |
| "logps/ref_chosen": -174.38418579101562, | |
| "logps/ref_rejected": -182.0985565185547, | |
| "logps/rejected": -183.1609649658203, | |
| "loss": 0.6928, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.009813489392399788, | |
| "rewards/margins": 0.0008107352769002318, | |
| "rewards/rejected": -0.010624224320054054, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13174145738987236, | |
| "grad_norm": 0.3784727156162262, | |
| "learning_rate": 4.989014955054745e-07, | |
| "logits/chosen": 10.471797943115234, | |
| "logits/rejected": 10.57437515258789, | |
| "logps/chosen": -173.78768920898438, | |
| "logps/ref_chosen": -172.48837280273438, | |
| "logps/ref_rejected": -179.625, | |
| "logps/rejected": -180.91595458984375, | |
| "loss": 0.6932, | |
| "rewards/accuracies": 0.4765625, | |
| "rewards/chosen": -0.012993087992072105, | |
| "rewards/margins": -8.335959864780307e-05, | |
| "rewards/rejected": -0.012909728102385998, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14491560312885962, | |
| "grad_norm": 0.4041476547718048, | |
| "learning_rate": 4.975306286336627e-07, | |
| "logits/chosen": 10.539186477661133, | |
| "logits/rejected": 10.477883338928223, | |
| "logps/chosen": -172.62416076660156, | |
| "logps/ref_chosen": -169.97216796875, | |
| "logps/ref_rejected": -180.00955200195312, | |
| "logps/rejected": -182.72525024414062, | |
| "loss": 0.6929, | |
| "rewards/accuracies": 0.546875, | |
| "rewards/chosen": -0.026519589126110077, | |
| "rewards/margins": 0.0006374535150825977, | |
| "rewards/rejected": -0.027157040312886238, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.15808974886784685, | |
| "grad_norm": 0.5964677333831787, | |
| "learning_rate": 4.956156357188939e-07, | |
| "logits/chosen": 10.28010082244873, | |
| "logits/rejected": 10.401304244995117, | |
| "logps/chosen": -178.43714904785156, | |
| "logps/ref_chosen": -175.83639526367188, | |
| "logps/ref_rejected": -181.73045349121094, | |
| "logps/rejected": -184.20223999023438, | |
| "loss": 0.6938, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.026007818058133125, | |
| "rewards/margins": -0.0012897354317829013, | |
| "rewards/rejected": -0.024718083441257477, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.17126389460683408, | |
| "grad_norm": 0.488361120223999, | |
| "learning_rate": 4.931607263312032e-07, | |
| "logits/chosen": 10.292243957519531, | |
| "logits/rejected": 10.386707305908203, | |
| "logps/chosen": -173.99058532714844, | |
| "logps/ref_chosen": -170.8806610107422, | |
| "logps/ref_rejected": -180.6468963623047, | |
| "logps/rejected": -184.01853942871094, | |
| "loss": 0.6919, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.03109920769929886, | |
| "rewards/margins": 0.002617327030748129, | |
| "rewards/rejected": -0.033716537058353424, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1844380403458213, | |
| "grad_norm": 0.45636966824531555, | |
| "learning_rate": 4.9017129689421e-07, | |
| "logits/chosen": 10.729025840759277, | |
| "logits/rejected": 10.921786308288574, | |
| "logps/chosen": -174.65345764160156, | |
| "logps/ref_chosen": -171.15963745117188, | |
| "logps/ref_rejected": -180.73471069335938, | |
| "logps/rejected": -184.30807495117188, | |
| "loss": 0.6928, | |
| "rewards/accuracies": 0.515625, | |
| "rewards/chosen": -0.03493800014257431, | |
| "rewards/margins": 0.0007955244509503245, | |
| "rewards/rejected": -0.03573352098464966, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.19761218608480857, | |
| "grad_norm": 0.3838670551776886, | |
| "learning_rate": 4.866539188226085e-07, | |
| "logits/chosen": 10.265353202819824, | |
| "logits/rejected": 10.378411293029785, | |
| "logps/chosen": -177.72991943359375, | |
| "logps/ref_chosen": -173.84222412109375, | |
| "logps/ref_rejected": -182.6996612548828, | |
| "logps/rejected": -186.89825439453125, | |
| "loss": 0.6916, | |
| "rewards/accuracies": 0.515625, | |
| "rewards/chosen": -0.03887684643268585, | |
| "rewards/margins": 0.003108714707195759, | |
| "rewards/rejected": -0.04198555648326874, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2107863318237958, | |
| "grad_norm": 0.3698700964450836, | |
| "learning_rate": 4.826163240767716e-07, | |
| "logits/chosen": 10.751184463500977, | |
| "logits/rejected": 10.945128440856934, | |
| "logps/chosen": -181.60301208496094, | |
| "logps/ref_chosen": -178.638671875, | |
| "logps/ref_rejected": -185.8243408203125, | |
| "logps/rejected": -188.93057250976562, | |
| "loss": 0.6925, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.029643364250659943, | |
| "rewards/margins": 0.0014189946232363582, | |
| "rewards/rejected": -0.03106236271560192, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.22396047756278303, | |
| "grad_norm": 0.5419145822525024, | |
| "learning_rate": 4.780673881662242e-07, | |
| "logits/chosen": 10.279337882995605, | |
| "logits/rejected": 10.265542984008789, | |
| "logps/chosen": -176.72821044921875, | |
| "logps/ref_chosen": -172.87332153320312, | |
| "logps/ref_rejected": -177.30355834960938, | |
| "logps/rejected": -181.23831176757812, | |
| "loss": 0.6928, | |
| "rewards/accuracies": 0.484375, | |
| "rewards/chosen": -0.038549020886421204, | |
| "rewards/margins": 0.0007985997945070267, | |
| "rewards/rejected": -0.03934762626886368, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.23713462330177026, | |
| "grad_norm": 0.5288915038108826, | |
| "learning_rate": 4.730171106393466e-07, | |
| "logits/chosen": 10.508716583251953, | |
| "logits/rejected": 10.601346969604492, | |
| "logps/chosen": -176.69735717773438, | |
| "logps/ref_chosen": -172.82644653320312, | |
| "logps/ref_rejected": -180.93768310546875, | |
| "logps/rejected": -184.82861328125, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.038708824664354324, | |
| "rewards/margins": 0.00020039017545059323, | |
| "rewards/rejected": -0.03890921175479889, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2503087690407575, | |
| "grad_norm": 0.45863935351371765, | |
| "learning_rate": 4.6747659310219757e-07, | |
| "logits/chosen": 10.509140014648438, | |
| "logits/rejected": 10.403264999389648, | |
| "logps/chosen": -183.62411499023438, | |
| "logps/ref_chosen": -179.5069122314453, | |
| "logps/ref_rejected": -185.01487731933594, | |
| "logps/rejected": -189.2952880859375, | |
| "loss": 0.6924, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.041171781718730927, | |
| "rewards/margins": 0.0016320659779012203, | |
| "rewards/rejected": -0.04280385002493858, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2634829147797447, | |
| "grad_norm": 0.6024115085601807, | |
| "learning_rate": 4.6145801481477433e-07, | |
| "logits/chosen": 10.452649116516113, | |
| "logits/rejected": 10.525249481201172, | |
| "logps/chosen": -174.3675994873047, | |
| "logps/ref_chosen": -170.23623657226562, | |
| "logps/ref_rejected": -177.55618286132812, | |
| "logps/rejected": -181.85946655273438, | |
| "loss": 0.6923, | |
| "rewards/accuracies": 0.5078125, | |
| "rewards/chosen": -0.04131368175148964, | |
| "rewards/margins": 0.0017191548831760883, | |
| "rewards/rejected": -0.04303283616900444, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.276657060518732, | |
| "grad_norm": 0.4595341384410858, | |
| "learning_rate": 4.549746059183561e-07, | |
| "logits/chosen": 10.422194480895996, | |
| "logits/rejected": 10.483356475830078, | |
| "logps/chosen": -180.44906616210938, | |
| "logps/ref_chosen": -176.3700408935547, | |
| "logps/ref_rejected": -182.80032348632812, | |
| "logps/rejected": -187.17276000976562, | |
| "loss": 0.6917, | |
| "rewards/accuracies": 0.5703125, | |
| "rewards/chosen": -0.04079030081629753, | |
| "rewards/margins": 0.0029342519119381905, | |
| "rewards/rejected": -0.04372455179691315, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.28983120625771924, | |
| "grad_norm": 0.3111652731895447, | |
| "learning_rate": 4.480406183527823e-07, | |
| "logits/chosen": 10.481276512145996, | |
| "logits/rejected": 10.559721946716309, | |
| "logps/chosen": -177.63865661621094, | |
| "logps/ref_chosen": -173.29742431640625, | |
| "logps/ref_rejected": -177.54916381835938, | |
| "logps/rejected": -182.2364044189453, | |
| "loss": 0.6915, | |
| "rewards/accuracies": 0.6015625, | |
| "rewards/chosen": -0.043412309139966965, | |
| "rewards/margins": 0.003460067557170987, | |
| "rewards/rejected": -0.04687237739562988, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.3030053519967065, | |
| "grad_norm": 0.520114541053772, | |
| "learning_rate": 4.4067129452759546e-07, | |
| "logits/chosen": 10.153634071350098, | |
| "logits/rejected": 10.433453559875488, | |
| "logps/chosen": -174.9644012451172, | |
| "logps/ref_chosen": -170.41177368164062, | |
| "logps/ref_rejected": -182.7073974609375, | |
| "logps/rejected": -187.5763397216797, | |
| "loss": 0.6916, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.04552610218524933, | |
| "rewards/margins": 0.0031630881130695343, | |
| "rewards/rejected": -0.04868919029831886, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3161794977356937, | |
| "grad_norm": 0.38157743215560913, | |
| "learning_rate": 4.3288283381591725e-07, | |
| "logits/chosen": 10.212328910827637, | |
| "logits/rejected": 10.149028778076172, | |
| "logps/chosen": -178.75759887695312, | |
| "logps/ref_chosen": -174.15003967285156, | |
| "logps/ref_rejected": -181.65594482421875, | |
| "logps/rejected": -186.58934020996094, | |
| "loss": 0.6916, | |
| "rewards/accuracies": 0.5703125, | |
| "rewards/chosen": -0.046075694262981415, | |
| "rewards/margins": 0.003258442971855402, | |
| "rewards/rejected": -0.04933413863182068, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.32935364347468093, | |
| "grad_norm": 0.5030648112297058, | |
| "learning_rate": 4.246923569447104e-07, | |
| "logits/chosen": 10.425580978393555, | |
| "logits/rejected": 10.567748069763184, | |
| "logps/chosen": -181.23281860351562, | |
| "logps/ref_chosen": -177.0232391357422, | |
| "logps/ref_rejected": -184.0133056640625, | |
| "logps/rejected": -188.3357696533203, | |
| "loss": 0.6926, | |
| "rewards/accuracies": 0.5234375, | |
| "rewards/chosen": -0.04209586977958679, | |
| "rewards/margins": 0.0011287340894341469, | |
| "rewards/rejected": -0.04322460666298866, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.34252778921366817, | |
| "grad_norm": 0.3347052335739136, | |
| "learning_rate": 4.161178683597054e-07, | |
| "logits/chosen": 10.395452499389648, | |
| "logits/rejected": 10.5776948928833, | |
| "logps/chosen": -177.6590576171875, | |
| "logps/ref_chosen": -172.2715606689453, | |
| "logps/ref_rejected": -180.46673583984375, | |
| "logps/rejected": -186.09349060058594, | |
| "loss": 0.692, | |
| "rewards/accuracies": 0.5703125, | |
| "rewards/chosen": -0.05387478694319725, | |
| "rewards/margins": 0.002392976311966777, | |
| "rewards/rejected": -0.056267764419317245, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3557019349526554, | |
| "grad_norm": 0.40187934041023254, | |
| "learning_rate": 4.0717821664772124e-07, | |
| "logits/chosen": 10.416847229003906, | |
| "logits/rejected": 10.693917274475098, | |
| "logps/chosen": -180.66729736328125, | |
| "logps/ref_chosen": -174.87091064453125, | |
| "logps/ref_rejected": -183.6770477294922, | |
| "logps/rejected": -189.53399658203125, | |
| "loss": 0.6929, | |
| "rewards/accuracies": 0.4921875, | |
| "rewards/chosen": -0.057963769882917404, | |
| "rewards/margins": 0.0006057576974853873, | |
| "rewards/rejected": -0.0585695244371891, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3688760806916426, | |
| "grad_norm": 0.537835419178009, | |
| "learning_rate": 3.978930531033806e-07, | |
| "logits/chosen": 10.315240859985352, | |
| "logits/rejected": 10.52540111541748, | |
| "logps/chosen": -178.289794921875, | |
| "logps/ref_chosen": -172.75225830078125, | |
| "logps/ref_rejected": -180.3443145751953, | |
| "logps/rejected": -186.29811096191406, | |
| "loss": 0.6911, | |
| "rewards/accuracies": 0.578125, | |
| "rewards/chosen": -0.05537542328238487, | |
| "rewards/margins": 0.004162484314292669, | |
| "rewards/rejected": -0.05953790992498398, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.3820502264306299, | |
| "grad_norm": 0.9168035984039307, | |
| "learning_rate": 3.882827885312998e-07, | |
| "logits/chosen": 10.197809219360352, | |
| "logits/rejected": 10.199935913085938, | |
| "logps/chosen": -181.69606018066406, | |
| "logps/ref_chosen": -176.17613220214844, | |
| "logps/ref_rejected": -180.6341552734375, | |
| "logps/rejected": -186.46124267578125, | |
| "loss": 0.6917, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.05519918352365494, | |
| "rewards/margins": 0.0030717107001692057, | |
| "rewards/rejected": -0.0582708939909935, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.39522437216961714, | |
| "grad_norm": 0.3577282428741455, | |
| "learning_rate": 3.7836854837871044e-07, | |
| "logits/chosen": 10.497368812561035, | |
| "logits/rejected": 10.725122451782227, | |
| "logps/chosen": -184.75241088867188, | |
| "logps/ref_chosen": -178.8612060546875, | |
| "logps/ref_rejected": -187.1546173095703, | |
| "logps/rejected": -193.5741424560547, | |
| "loss": 0.6906, | |
| "rewards/accuracies": 0.5859375, | |
| "rewards/chosen": -0.058911994099617004, | |
| "rewards/margins": 0.0052833897061645985, | |
| "rewards/rejected": -0.06419539451599121, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4083985179086044, | |
| "grad_norm": 0.46136385202407837, | |
| "learning_rate": 3.681721262971413e-07, | |
| "logits/chosen": 10.34453010559082, | |
| "logits/rejected": 10.35668659210205, | |
| "logps/chosen": -178.43930053710938, | |
| "logps/ref_chosen": -172.60848999023438, | |
| "logps/ref_rejected": -179.45013427734375, | |
| "logps/rejected": -185.967529296875, | |
| "loss": 0.6898, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.0583079531788826, | |
| "rewards/margins": 0.006865999195724726, | |
| "rewards/rejected": -0.06517395377159119, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.4215726636475916, | |
| "grad_norm": 0.3592880368232727, | |
| "learning_rate": 3.577159362352426e-07, | |
| "logits/chosen": 10.357152938842773, | |
| "logits/rejected": 10.40371322631836, | |
| "logps/chosen": -181.8780975341797, | |
| "logps/ref_chosen": -175.29119873046875, | |
| "logps/ref_rejected": -182.2250213623047, | |
| "logps/rejected": -189.01063537597656, | |
| "loss": 0.6923, | |
| "rewards/accuracies": 0.5390625, | |
| "rewards/chosen": -0.06586895883083344, | |
| "rewards/margins": 0.001987436320632696, | |
| "rewards/rejected": -0.06785639375448227, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.43474680938657884, | |
| "grad_norm": 0.4338357746601105, | |
| "learning_rate": 3.470229631680624e-07, | |
| "logits/chosen": 10.255874633789062, | |
| "logits/rejected": 10.296948432922363, | |
| "logps/chosen": -183.95484924316406, | |
| "logps/ref_chosen": -177.42674255371094, | |
| "logps/ref_rejected": -181.18801879882812, | |
| "logps/rejected": -188.07952880859375, | |
| "loss": 0.6914, | |
| "rewards/accuracies": 0.5859375, | |
| "rewards/chosen": -0.065280981361866, | |
| "rewards/margins": 0.003634275868535042, | |
| "rewards/rejected": -0.06891525536775589, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.44792095512556607, | |
| "grad_norm": 0.4542549252510071, | |
| "learning_rate": 3.361167125710832e-07, | |
| "logits/chosen": 10.056829452514648, | |
| "logits/rejected": 10.121377944946289, | |
| "logps/chosen": -181.4767303466797, | |
| "logps/ref_chosen": -173.72467041015625, | |
| "logps/ref_rejected": -180.5912322998047, | |
| "logps/rejected": -188.53164672851562, | |
| "loss": 0.6923, | |
| "rewards/accuracies": 0.4921875, | |
| "rewards/chosen": -0.07752064615488052, | |
| "rewards/margins": 0.0018832057248800993, | |
| "rewards/rejected": -0.0794038400053978, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.4610951008645533, | |
| "grad_norm": 0.4729287028312683, | |
| "learning_rate": 3.2502115875008516e-07, | |
| "logits/chosen": 10.46947193145752, | |
| "logits/rejected": 10.667445182800293, | |
| "logps/chosen": -182.939208984375, | |
| "logps/ref_chosen": -175.0980987548828, | |
| "logps/ref_rejected": -181.65821838378906, | |
| "logps/rejected": -189.89309692382812, | |
| "loss": 0.6913, | |
| "rewards/accuracies": 0.578125, | |
| "rewards/chosen": -0.07841099798679352, | |
| "rewards/margins": 0.00393773103132844, | |
| "rewards/rejected": -0.08234872668981552, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.47426924660354053, | |
| "grad_norm": 0.8570416569709778, | |
| "learning_rate": 3.137606921404191e-07, | |
| "logits/chosen": 10.280503273010254, | |
| "logits/rejected": 10.623340606689453, | |
| "logps/chosen": -182.2921142578125, | |
| "logps/ref_chosen": -174.0312042236328, | |
| "logps/ref_rejected": -181.84324645996094, | |
| "logps/rejected": -190.1889190673828, | |
| "loss": 0.6928, | |
| "rewards/accuracies": 0.484375, | |
| "rewards/chosen": -0.08260920643806458, | |
| "rewards/margins": 0.0008472882909700274, | |
| "rewards/rejected": -0.08345648646354675, | |
| "step": 36 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 12, | |
| "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 | |
| } | |