| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9950825430277486, | |
| "eval_steps": 10000, | |
| "global_step": 710, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.4084507042253522e-07, | |
| "logits/chosen": -0.024068962782621384, | |
| "logits/rejected": 0.04506533965468407, | |
| "logps/chosen": -317.1720275878906, | |
| "logps/rejected": -207.3963623046875, | |
| "loss": 0.6933, | |
| "rewards/accuracies": 0.4124999940395355, | |
| "rewards/chosen": -0.0009543737396597862, | |
| "rewards/margins": -0.0014725492801517248, | |
| "rewards/rejected": 0.0005181756569072604, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.8169014084507043e-07, | |
| "logits/chosen": -0.04674551263451576, | |
| "logits/rejected": -0.010908829048275948, | |
| "logps/chosen": -294.0361328125, | |
| "logps/rejected": -201.79061889648438, | |
| "loss": 0.6914, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": 0.0003093578852713108, | |
| "rewards/margins": 0.004586204886436462, | |
| "rewards/rejected": -0.004276847001165152, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.225352112676056e-07, | |
| "logits/chosen": -0.04512999951839447, | |
| "logits/rejected": 0.007197662256658077, | |
| "logps/chosen": -361.146240234375, | |
| "logps/rejected": -249.73080444335938, | |
| "loss": 0.6841, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.001304158358834684, | |
| "rewards/margins": 0.015663906931877136, | |
| "rewards/rejected": -0.01696806587278843, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5.633802816901409e-07, | |
| "logits/chosen": -0.04949279874563217, | |
| "logits/rejected": 0.00024662315263412893, | |
| "logps/chosen": -291.42913818359375, | |
| "logps/rejected": -196.7120361328125, | |
| "loss": 0.6759, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.015403245575726032, | |
| "rewards/margins": 0.05249834060668945, | |
| "rewards/rejected": -0.06790158897638321, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 7.04225352112676e-07, | |
| "logits/chosen": 0.02841530181467533, | |
| "logits/rejected": 0.0736595019698143, | |
| "logps/chosen": -373.4976806640625, | |
| "logps/rejected": -227.70068359375, | |
| "loss": 0.665, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -0.05390547588467598, | |
| "rewards/margins": 0.11358483880758286, | |
| "rewards/rejected": -0.16749031841754913, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.450704225352112e-07, | |
| "logits/chosen": -0.03540544956922531, | |
| "logits/rejected": 0.025740886107087135, | |
| "logps/chosen": -323.0583801269531, | |
| "logps/rejected": -230.3257293701172, | |
| "loss": 0.6567, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.03229106217622757, | |
| "rewards/margins": 0.1074444055557251, | |
| "rewards/rejected": -0.13973546028137207, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.859154929577465e-07, | |
| "logits/chosen": -0.06292358785867691, | |
| "logits/rejected": -0.009262708015739918, | |
| "logps/chosen": -354.81341552734375, | |
| "logps/rejected": -255.65560913085938, | |
| "loss": 0.6477, | |
| "rewards/accuracies": 0.6187499761581421, | |
| "rewards/chosen": -0.03358080983161926, | |
| "rewards/margins": 0.17669352889060974, | |
| "rewards/rejected": -0.210274338722229, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.995106132599868e-07, | |
| "logits/chosen": 0.03767090290784836, | |
| "logits/rejected": 0.09468533098697662, | |
| "logps/chosen": -319.8956298828125, | |
| "logps/rejected": -235.5646514892578, | |
| "loss": 0.667, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.07292340695858002, | |
| "rewards/margins": 0.1298426389694214, | |
| "rewards/rejected": -0.2027660608291626, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.978201358980644e-07, | |
| "logits/chosen": -0.016046693548560143, | |
| "logits/rejected": 0.04123395308852196, | |
| "logps/chosen": -361.1316223144531, | |
| "logps/rejected": -279.39520263671875, | |
| "loss": 0.6491, | |
| "rewards/accuracies": 0.625, | |
| "rewards/chosen": 0.030158694833517075, | |
| "rewards/margins": 0.1393767148256302, | |
| "rewards/rejected": -0.10921802371740341, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.949266103908894e-07, | |
| "logits/chosen": -0.01002789568156004, | |
| "logits/rejected": 0.055082082748413086, | |
| "logps/chosen": -275.36773681640625, | |
| "logps/rejected": -211.86721801757812, | |
| "loss": 0.6472, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": 0.04291190952062607, | |
| "rewards/margins": 0.11093775182962418, | |
| "rewards/rejected": -0.06802584230899811, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.908370293252287e-07, | |
| "logits/chosen": -0.008449924178421497, | |
| "logits/rejected": 0.018720177933573723, | |
| "logps/chosen": -323.53802490234375, | |
| "logps/rejected": -244.490966796875, | |
| "loss": 0.6407, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": 0.07049892842769623, | |
| "rewards/margins": 0.1456976681947708, | |
| "rewards/rejected": -0.07519873231649399, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 9.855612757141654e-07, | |
| "logits/chosen": -0.07508296519517899, | |
| "logits/rejected": -0.012953217141330242, | |
| "logps/chosen": -336.1876525878906, | |
| "logps/rejected": -208.52011108398438, | |
| "loss": 0.6333, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.03278186544775963, | |
| "rewards/margins": 0.26016712188720703, | |
| "rewards/rejected": -0.2273852527141571, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.791120991134902e-07, | |
| "logits/chosen": -0.10753818601369858, | |
| "logits/rejected": -0.01082658488303423, | |
| "logps/chosen": -376.611083984375, | |
| "logps/rejected": -251.7242889404297, | |
| "loss": 0.6491, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": -0.026872599497437477, | |
| "rewards/margins": 0.2888794541358948, | |
| "rewards/rejected": -0.3157520890235901, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.715050848107168e-07, | |
| "logits/chosen": -0.17280681431293488, | |
| "logits/rejected": -0.0874427855014801, | |
| "logps/chosen": -342.14593505859375, | |
| "logps/rejected": -229.78652954101562, | |
| "loss": 0.6346, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -0.014674955978989601, | |
| "rewards/margins": 0.2581936717033386, | |
| "rewards/rejected": -0.2728686034679413, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 9.627586161611731e-07, | |
| "logits/chosen": -0.09239596873521805, | |
| "logits/rejected": -0.035311758518218994, | |
| "logps/chosen": -282.2192077636719, | |
| "logps/rejected": -214.80810546875, | |
| "loss": 0.6501, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.03641432523727417, | |
| "rewards/margins": 0.19125035405158997, | |
| "rewards/rejected": -0.22766467928886414, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 9.528938301621955e-07, | |
| "logits/chosen": -0.168453648686409, | |
| "logits/rejected": -0.1073572039604187, | |
| "logps/chosen": -341.64599609375, | |
| "logps/rejected": -223.13101196289062, | |
| "loss": 0.6342, | |
| "rewards/accuracies": 0.6312500238418579, | |
| "rewards/chosen": 0.01668923906981945, | |
| "rewards/margins": 0.19806070625782013, | |
| "rewards/rejected": -0.18137145042419434, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 9.419345663727804e-07, | |
| "logits/chosen": -0.13118991255760193, | |
| "logits/rejected": -0.0702921599149704, | |
| "logps/chosen": -344.593505859375, | |
| "logps/rejected": -231.71804809570312, | |
| "loss": 0.6221, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.004311258438974619, | |
| "rewards/margins": 0.2375575602054596, | |
| "rewards/rejected": -0.23324629664421082, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 9.299073093021404e-07, | |
| "logits/chosen": -0.05861334875226021, | |
| "logits/rejected": 0.004674489144235849, | |
| "logps/chosen": -355.00872802734375, | |
| "logps/rejected": -255.2248992919922, | |
| "loss": 0.6285, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.13683992624282837, | |
| "rewards/margins": 0.26262304186820984, | |
| "rewards/rejected": -0.3994629979133606, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 9.168411244063861e-07, | |
| "logits/chosen": -0.15201464295387268, | |
| "logits/rejected": -0.08699367940425873, | |
| "logps/chosen": -369.53936767578125, | |
| "logps/rejected": -261.6515197753906, | |
| "loss": 0.6253, | |
| "rewards/accuracies": 0.637499988079071, | |
| "rewards/chosen": -0.18763788044452667, | |
| "rewards/margins": 0.2461237609386444, | |
| "rewards/rejected": -0.4337615966796875, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 9.02767587848013e-07, | |
| "logits/chosen": -0.1659875363111496, | |
| "logits/rejected": -0.11425626277923584, | |
| "logps/chosen": -322.26971435546875, | |
| "logps/rejected": -269.41571044921875, | |
| "loss": 0.6377, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.1545395851135254, | |
| "rewards/margins": 0.16578510403633118, | |
| "rewards/rejected": -0.32032471895217896, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.877207101879301e-07, | |
| "logits/chosen": -0.12081719934940338, | |
| "logits/rejected": -0.055270951241254807, | |
| "logps/chosen": -303.1000061035156, | |
| "logps/rejected": -226.3379364013672, | |
| "loss": 0.6265, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -0.1504133641719818, | |
| "rewards/margins": 0.2184509038925171, | |
| "rewards/rejected": -0.3688642680644989, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.717368541944452e-07, | |
| "logits/chosen": -0.14912042021751404, | |
| "logits/rejected": -0.11508848518133163, | |
| "logps/chosen": -313.95745849609375, | |
| "logps/rejected": -232.803955078125, | |
| "loss": 0.6245, | |
| "rewards/accuracies": 0.606249988079071, | |
| "rewards/chosen": -0.17989172041416168, | |
| "rewards/margins": 0.21268931031227112, | |
| "rewards/rejected": -0.3925810158252716, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 8.54854646967831e-07, | |
| "logits/chosen": -0.20850150287151337, | |
| "logits/rejected": -0.14839352667331696, | |
| "logps/chosen": -351.56768798828125, | |
| "logps/rejected": -286.0954284667969, | |
| "loss": 0.6232, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": -0.20482465624809265, | |
| "rewards/margins": 0.2566435933113098, | |
| "rewards/rejected": -0.46146830916404724, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 8.371148865928318e-07, | |
| "logits/chosen": -0.13414549827575684, | |
| "logits/rejected": -0.09452676773071289, | |
| "logps/chosen": -334.4315490722656, | |
| "logps/rejected": -271.9152526855469, | |
| "loss": 0.6104, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.18954138457775116, | |
| "rewards/margins": 0.24784322082996368, | |
| "rewards/rejected": -0.43738460540771484, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 8.185604435447001e-07, | |
| "logits/chosen": -0.08637334406375885, | |
| "logits/rejected": -0.03546728193759918, | |
| "logps/chosen": -349.38787841796875, | |
| "logps/rejected": -278.61761474609375, | |
| "loss": 0.6238, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.2417462319135666, | |
| "rewards/margins": 0.2916719615459442, | |
| "rewards/rejected": -0.5334181785583496, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 7.992361570870287e-07, | |
| "logits/chosen": -0.15081752836704254, | |
| "logits/rejected": -0.11529238522052765, | |
| "logps/chosen": -311.26507568359375, | |
| "logps/rejected": -253.03659057617188, | |
| "loss": 0.625, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": -0.34321683645248413, | |
| "rewards/margins": 0.19419686496257782, | |
| "rewards/rejected": -0.5374137163162231, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 7.791887269117441e-07, | |
| "logits/chosen": -0.08736265450716019, | |
| "logits/rejected": 0.010209694504737854, | |
| "logps/chosen": -350.77197265625, | |
| "logps/rejected": -249.4473114013672, | |
| "loss": 0.6198, | |
| "rewards/accuracies": 0.668749988079071, | |
| "rewards/chosen": -0.33235999941825867, | |
| "rewards/margins": 0.3289110064506531, | |
| "rewards/rejected": -0.6612709760665894, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 7.584666002831294e-07, | |
| "logits/chosen": -0.17468181252479553, | |
| "logits/rejected": -0.10544611513614655, | |
| "logps/chosen": -393.7838439941406, | |
| "logps/rejected": -275.46636962890625, | |
| "loss": 0.6284, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.20970232784748077, | |
| "rewards/margins": 0.31893718242645264, | |
| "rewards/rejected": -0.5286394357681274, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 7.37119854958609e-07, | |
| "logits/chosen": -0.18520912528038025, | |
| "logits/rejected": -0.12915001809597015, | |
| "logps/chosen": -372.53961181640625, | |
| "logps/rejected": -301.4205017089844, | |
| "loss": 0.6426, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.3479081690311432, | |
| "rewards/margins": 0.25431200861930847, | |
| "rewards/rejected": -0.6022201776504517, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 7.152000781692285e-07, | |
| "logits/chosen": -0.20307299494743347, | |
| "logits/rejected": -0.1757517158985138, | |
| "logps/chosen": -368.33392333984375, | |
| "logps/rejected": -302.0033874511719, | |
| "loss": 0.6219, | |
| "rewards/accuracies": 0.6499999761581421, | |
| "rewards/chosen": -0.2545304298400879, | |
| "rewards/margins": 0.2229403257369995, | |
| "rewards/rejected": -0.4774707853794098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.927602419522946e-07, | |
| "logits/chosen": -0.15039893984794617, | |
| "logits/rejected": -0.1383979469537735, | |
| "logps/chosen": -319.64349365234375, | |
| "logps/rejected": -265.85198974609375, | |
| "loss": 0.624, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.3667163848876953, | |
| "rewards/margins": 0.18039202690124512, | |
| "rewards/rejected": -0.5471083521842957, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 6.698545751374463e-07, | |
| "logits/chosen": -0.25044673681259155, | |
| "logits/rejected": -0.12538060545921326, | |
| "logps/chosen": -413.25946044921875, | |
| "logps/rejected": -303.199951171875, | |
| "loss": 0.6216, | |
| "rewards/accuracies": 0.6937500238418579, | |
| "rewards/chosen": -0.3125981390476227, | |
| "rewards/margins": 0.38619184494018555, | |
| "rewards/rejected": -0.6987899541854858, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.465384322955224e-07, | |
| "logits/chosen": -0.1757752150297165, | |
| "logits/rejected": -0.11078636348247528, | |
| "logps/chosen": -362.4681091308594, | |
| "logps/rejected": -267.5201416015625, | |
| "loss": 0.6074, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.33040475845336914, | |
| "rewards/margins": 0.37607377767562866, | |
| "rewards/rejected": -0.7064785957336426, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 6.228681599669248e-07, | |
| "logits/chosen": -0.19059506058692932, | |
| "logits/rejected": -0.09184812009334564, | |
| "logps/chosen": -430.0191345214844, | |
| "logps/rejected": -296.02557373046875, | |
| "loss": 0.617, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.3559810519218445, | |
| "rewards/margins": 0.3992285132408142, | |
| "rewards/rejected": -0.7552096247673035, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 5.989009604927586e-07, | |
| "logits/chosen": -0.1337263584136963, | |
| "logits/rejected": -0.01602357253432274, | |
| "logps/chosen": -391.5013122558594, | |
| "logps/rejected": -306.56982421875, | |
| "loss": 0.5985, | |
| "rewards/accuracies": 0.7124999761581421, | |
| "rewards/chosen": -0.3254554867744446, | |
| "rewards/margins": 0.4386405944824219, | |
| "rewards/rejected": -0.7640960812568665, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 5.74694753777815e-07, | |
| "logits/chosen": -0.13974535465240479, | |
| "logits/rejected": -0.11217441409826279, | |
| "logps/chosen": -327.7983703613281, | |
| "logps/rejected": -279.513916015625, | |
| "loss": 0.545, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": -0.42400461435317993, | |
| "rewards/margins": 0.4070449769496918, | |
| "rewards/rejected": -0.8310495615005493, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.503080373194666e-07, | |
| "logits/chosen": -0.2235272228717804, | |
| "logits/rejected": -0.12439509481191635, | |
| "logps/chosen": -374.24261474609375, | |
| "logps/rejected": -333.127685546875, | |
| "loss": 0.4827, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -0.4394102096557617, | |
| "rewards/margins": 0.6707636713981628, | |
| "rewards/rejected": -1.1101738214492798, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.257997448407366e-07, | |
| "logits/chosen": -0.12490369379520416, | |
| "logits/rejected": 0.020254041999578476, | |
| "logps/chosen": -408.45367431640625, | |
| "logps/rejected": -308.180908203125, | |
| "loss": 0.4897, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -0.4755324423313141, | |
| "rewards/margins": 0.6596079468727112, | |
| "rewards/rejected": -1.1351404190063477, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 5.012291038691665e-07, | |
| "logits/chosen": -0.15194548666477203, | |
| "logits/rejected": -0.03352439031004906, | |
| "logps/chosen": -419.17730712890625, | |
| "logps/rejected": -391.656982421875, | |
| "loss": 0.4746, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -0.6772257089614868, | |
| "rewards/margins": 0.8299336433410645, | |
| "rewards/rejected": -1.5071594715118408, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.7665549260567063e-07, | |
| "logits/chosen": -0.20493462681770325, | |
| "logits/rejected": -0.13634856045246124, | |
| "logps/chosen": -411.919189453125, | |
| "logps/rejected": -344.78759765625, | |
| "loss": 0.4713, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -0.7057473063468933, | |
| "rewards/margins": 0.824458122253418, | |
| "rewards/rejected": -1.530205249786377, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.521382964292663e-07, | |
| "logits/chosen": -0.14551517367362976, | |
| "logits/rejected": -0.07351706176996231, | |
| "logps/chosen": -394.6050109863281, | |
| "logps/rejected": -345.28082275390625, | |
| "loss": 0.4741, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": -0.6415594816207886, | |
| "rewards/margins": 0.7878659963607788, | |
| "rewards/rejected": -1.4294254779815674, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.277367643844574e-07, | |
| "logits/chosen": -0.1973930448293686, | |
| "logits/rejected": -0.10258449614048004, | |
| "logps/chosen": -446.4418029785156, | |
| "logps/rejected": -366.457763671875, | |
| "loss": 0.4605, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -0.8334900736808777, | |
| "rewards/margins": 0.7750234603881836, | |
| "rewards/rejected": -1.608513593673706, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.035098659980891e-07, | |
| "logits/chosen": -0.1986023634672165, | |
| "logits/rejected": -0.06475992500782013, | |
| "logps/chosen": -419.17010498046875, | |
| "logps/rejected": -395.53509521484375, | |
| "loss": 0.465, | |
| "rewards/accuracies": 0.8062499761581421, | |
| "rewards/chosen": -1.0268628597259521, | |
| "rewards/margins": 0.7243725657463074, | |
| "rewards/rejected": -1.7512352466583252, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.795161487716928e-07, | |
| "logits/chosen": -0.20193979144096375, | |
| "logits/rejected": -0.09314943850040436, | |
| "logps/chosen": -491.21405029296875, | |
| "logps/rejected": -423.76654052734375, | |
| "loss": 0.4524, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -1.0318418741226196, | |
| "rewards/margins": 0.9399174451828003, | |
| "rewards/rejected": -1.9717592000961304, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.5581359669371223e-07, | |
| "logits/chosen": -0.11288833618164062, | |
| "logits/rejected": -0.1004786491394043, | |
| "logps/chosen": -417.3036193847656, | |
| "logps/rejected": -372.6753845214844, | |
| "loss": 0.4561, | |
| "rewards/accuracies": 0.831250011920929, | |
| "rewards/chosen": -0.969109058380127, | |
| "rewards/margins": 0.6991706490516663, | |
| "rewards/rejected": -1.6682794094085693, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.324594901135326e-07, | |
| "logits/chosen": -0.17446324229240417, | |
| "logits/rejected": -0.0842059999704361, | |
| "logps/chosen": -438.217041015625, | |
| "logps/rejected": -390.52874755859375, | |
| "loss": 0.459, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -1.0395996570587158, | |
| "rewards/margins": 0.7797072529792786, | |
| "rewards/rejected": -1.8193069696426392, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.095102673159463e-07, | |
| "logits/chosen": -0.13965001702308655, | |
| "logits/rejected": -0.09614251554012299, | |
| "logps/chosen": -477.21337890625, | |
| "logps/rejected": -420.8619079589844, | |
| "loss": 0.4336, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": -1.1968367099761963, | |
| "rewards/margins": 0.8374635577201843, | |
| "rewards/rejected": -2.0343000888824463, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.870213881305802e-07, | |
| "logits/chosen": -0.06153956800699234, | |
| "logits/rejected": 0.04337477311491966, | |
| "logps/chosen": -457.060302734375, | |
| "logps/rejected": -406.51300048828125, | |
| "loss": 0.4411, | |
| "rewards/accuracies": 0.793749988079071, | |
| "rewards/chosen": -1.1784467697143555, | |
| "rewards/margins": 0.8234208226203918, | |
| "rewards/rejected": -2.0018675327301025, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.6504719990588745e-07, | |
| "logits/chosen": -0.05881907790899277, | |
| "logits/rejected": 0.013253748416900635, | |
| "logps/chosen": -455.1761169433594, | |
| "logps/rejected": -401.30133056640625, | |
| "loss": 0.4396, | |
| "rewards/accuracies": 0.831250011920929, | |
| "rewards/chosen": -1.2112154960632324, | |
| "rewards/margins": 0.8642939329147339, | |
| "rewards/rejected": -2.0755093097686768, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.436408061715988e-07, | |
| "logits/chosen": 0.04037608206272125, | |
| "logits/rejected": 0.047458432614803314, | |
| "logps/chosen": -382.39141845703125, | |
| "logps/rejected": -422.6522521972656, | |
| "loss": 0.4409, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": -1.066866159439087, | |
| "rewards/margins": 0.8686450719833374, | |
| "rewards/rejected": -1.9355109930038452, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.22853938307025e-07, | |
| "logits/chosen": 0.09550214558839798, | |
| "logits/rejected": 0.1434255838394165, | |
| "logps/chosen": -393.2830810546875, | |
| "logps/rejected": -355.2184753417969, | |
| "loss": 0.4469, | |
| "rewards/accuracies": 0.8187500238418579, | |
| "rewards/chosen": -1.153294324874878, | |
| "rewards/margins": 0.6956819891929626, | |
| "rewards/rejected": -1.8489763736724854, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.0273683052534173e-07, | |
| "logits/chosen": 0.14489802718162537, | |
| "logits/rejected": 0.22391514480113983, | |
| "logps/chosen": -465.29595947265625, | |
| "logps/rejected": -435.1458435058594, | |
| "loss": 0.4414, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -1.2600550651550293, | |
| "rewards/margins": 0.9257229566574097, | |
| "rewards/rejected": -2.1857781410217285, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 1.833380984759764e-07, | |
| "logits/chosen": 0.07251317799091339, | |
| "logits/rejected": 0.17206279933452606, | |
| "logps/chosen": -410.1485900878906, | |
| "logps/rejected": -438.0814514160156, | |
| "loss": 0.4487, | |
| "rewards/accuracies": 0.8062499761581421, | |
| "rewards/chosen": -1.2980937957763672, | |
| "rewards/margins": 0.8264884948730469, | |
| "rewards/rejected": -2.124582290649414, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1.6470462175846606e-07, | |
| "logits/chosen": 0.09354039281606674, | |
| "logits/rejected": 0.13454324007034302, | |
| "logps/chosen": -464.46990966796875, | |
| "logps/rejected": -444.4037170410156, | |
| "loss": 0.4383, | |
| "rewards/accuracies": 0.793749988079071, | |
| "rewards/chosen": -1.0875951051712036, | |
| "rewards/margins": 0.913548469543457, | |
| "rewards/rejected": -2.001143455505371, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.468814306317092e-07, | |
| "logits/chosen": 0.12879224121570587, | |
| "logits/rejected": 0.1979069709777832, | |
| "logps/chosen": -426.343505859375, | |
| "logps/rejected": -380.24493408203125, | |
| "loss": 0.4398, | |
| "rewards/accuracies": 0.856249988079071, | |
| "rewards/chosen": -1.0752108097076416, | |
| "rewards/margins": 0.9150044322013855, | |
| "rewards/rejected": -1.9902150630950928, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.299115971923958e-07, | |
| "logits/chosen": 0.12415747344493866, | |
| "logits/rejected": 0.20912082493305206, | |
| "logps/chosen": -448.64031982421875, | |
| "logps/rejected": -442.22576904296875, | |
| "loss": 0.4317, | |
| "rewards/accuracies": 0.8687499761581421, | |
| "rewards/chosen": -1.2348341941833496, | |
| "rewards/margins": 0.9574505090713501, | |
| "rewards/rejected": -2.1922848224639893, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.1383613128559305e-07, | |
| "logits/chosen": 0.024107109755277634, | |
| "logits/rejected": 0.1119920164346695, | |
| "logps/chosen": -505.82928466796875, | |
| "logps/rejected": -505.4091796875, | |
| "loss": 0.426, | |
| "rewards/accuracies": 0.8687499761581421, | |
| "rewards/chosen": -1.2975566387176514, | |
| "rewards/margins": 1.085707187652588, | |
| "rewards/rejected": -2.3832640647888184, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.869388139903495e-08, | |
| "logits/chosen": 0.012987576425075531, | |
| "logits/rejected": 0.06353282183408737, | |
| "logps/chosen": -541.9244995117188, | |
| "logps/rejected": -499.44110107421875, | |
| "loss": 0.4104, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -1.3797237873077393, | |
| "rewards/margins": 0.9557130932807922, | |
| "rewards/rejected": -2.335437059402466, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 8.452144078061818e-08, | |
| "logits/chosen": 0.14748625457286835, | |
| "logits/rejected": 0.21033573150634766, | |
| "logps/chosen": -428.7500915527344, | |
| "logps/rejected": -418.6878356933594, | |
| "loss": 0.4248, | |
| "rewards/accuracies": 0.8062499761581421, | |
| "rewards/chosen": -1.361243724822998, | |
| "rewards/margins": 0.893297553062439, | |
| "rewards/rejected": -2.2545411586761475, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 7.135305900598321e-08, | |
| "logits/chosen": 0.012422094121575356, | |
| "logits/rejected": 0.08027663081884384, | |
| "logps/chosen": -506.3257751464844, | |
| "logps/rejected": -460.52777099609375, | |
| "loss": 0.4245, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -1.5179449319839478, | |
| "rewards/margins": 1.0011281967163086, | |
| "rewards/rejected": -2.5190727710723877, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 5.9220559209888166e-08, | |
| "logits/chosen": 0.12760373950004578, | |
| "logits/rejected": 0.20078882575035095, | |
| "logps/chosen": -452.77734375, | |
| "logps/rejected": -487.36016845703125, | |
| "loss": 0.4266, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -1.444874882698059, | |
| "rewards/margins": 1.0283567905426025, | |
| "rewards/rejected": -2.473231792449951, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 4.815326118139812e-08, | |
| "logits/chosen": 0.14059333503246307, | |
| "logits/rejected": 0.2606312930583954, | |
| "logps/chosen": -415.8949279785156, | |
| "logps/rejected": -418.4481506347656, | |
| "loss": 0.4339, | |
| "rewards/accuracies": 0.8187500238418579, | |
| "rewards/chosen": -1.5133248567581177, | |
| "rewards/margins": 0.8190656900405884, | |
| "rewards/rejected": -2.332390546798706, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.81779105087407e-08, | |
| "logits/chosen": 0.08177526295185089, | |
| "logits/rejected": 0.21824178099632263, | |
| "logps/chosen": -476.4537658691406, | |
| "logps/rejected": -467.0603942871094, | |
| "loss": 0.4436, | |
| "rewards/accuracies": 0.8125, | |
| "rewards/chosen": -1.559093713760376, | |
| "rewards/margins": 0.9111778140068054, | |
| "rewards/rejected": -2.470271587371826, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.9318613945057637e-08, | |
| "logits/chosen": 0.12043829262256622, | |
| "logits/rejected": 0.19589121639728546, | |
| "logps/chosen": -504.67608642578125, | |
| "logps/rejected": -489.88330078125, | |
| "loss": 0.4181, | |
| "rewards/accuracies": 0.856249988079071, | |
| "rewards/chosen": -1.4812028408050537, | |
| "rewards/margins": 1.078580617904663, | |
| "rewards/rejected": -2.559783458709717, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 2.1596781151249523e-08, | |
| "logits/chosen": 0.15832188725471497, | |
| "logits/rejected": 0.2765315771102905, | |
| "logps/chosen": -461.1602478027344, | |
| "logps/rejected": -456.65826416015625, | |
| "loss": 0.4321, | |
| "rewards/accuracies": 0.8187500238418579, | |
| "rewards/chosen": -1.4598538875579834, | |
| "rewards/margins": 0.8676374554634094, | |
| "rewards/rejected": -2.327491283416748, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.5031072956701695e-08, | |
| "logits/chosen": 0.13437607884407043, | |
| "logits/rejected": 0.19320687651634216, | |
| "logps/chosen": -503.2486267089844, | |
| "logps/rejected": -498.3321228027344, | |
| "loss": 0.4196, | |
| "rewards/accuracies": 0.856249988079071, | |
| "rewards/chosen": -1.5257575511932373, | |
| "rewards/margins": 1.106710433959961, | |
| "rewards/rejected": -2.632467746734619, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 9.637356262923723e-09, | |
| "logits/chosen": 0.24336127936840057, | |
| "logits/rejected": 0.2697839140892029, | |
| "logps/chosen": -468.03289794921875, | |
| "logps/rejected": -444.1060485839844, | |
| "loss": 0.4387, | |
| "rewards/accuracies": 0.856249988079071, | |
| "rewards/chosen": -1.5293312072753906, | |
| "rewards/margins": 0.842644214630127, | |
| "rewards/rejected": -2.3719754219055176, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 5.428665699084789e-09, | |
| "logits/chosen": 0.10800081491470337, | |
| "logits/rejected": 0.1402808427810669, | |
| "logps/chosen": -492.41448974609375, | |
| "logps/rejected": -494.878173828125, | |
| "loss": 0.4088, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": -1.4665706157684326, | |
| "rewards/margins": 1.0468090772628784, | |
| "rewards/rejected": -2.5133795738220215, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 2.415172122110343e-09, | |
| "logits/chosen": 0.13986334204673767, | |
| "logits/rejected": 0.26917481422424316, | |
| "logps/chosen": -483.06024169921875, | |
| "logps/rejected": -482.5399475097656, | |
| "loss": 0.4382, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": -1.4039775133132935, | |
| "rewards/margins": 1.0037094354629517, | |
| "rewards/rejected": -2.407686948776245, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 6.041580374618327e-10, | |
| "logits/chosen": 0.105182945728302, | |
| "logits/rejected": 0.17750731110572815, | |
| "logps/chosen": -485.490478515625, | |
| "logps/rejected": -456.46063232421875, | |
| "loss": 0.4411, | |
| "rewards/accuracies": 0.84375, | |
| "rewards/chosen": -1.5071966648101807, | |
| "rewards/margins": 0.8808605074882507, | |
| "rewards/rejected": -2.388056993484497, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0, | |
| "logits/chosen": 0.1599053144454956, | |
| "logits/rejected": 0.2173675298690796, | |
| "logps/chosen": -413.2229919433594, | |
| "logps/rejected": -438.8427734375, | |
| "loss": 0.4277, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -1.4848761558532715, | |
| "rewards/margins": 0.8370813131332397, | |
| "rewards/rejected": -2.3219573497772217, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 710, | |
| "total_flos": 0.0, | |
| "train_loss": 0.5407548575334146, | |
| "train_runtime": 10382.7821, | |
| "train_samples_per_second": 8.774, | |
| "train_steps_per_second": 0.068 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 710, | |
| "num_train_epochs": 2, | |
| "save_steps": 10000, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |