{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.998691442030882, "eval_steps": 500, "global_step": 477, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.010468463752944255, "grad_norm": 30.248635446936238, "learning_rate": 5.208333333333333e-08, "logits/chosen": -0.4978736937046051, "logits/rejected": -0.5136879682540894, "logps/chosen": -1.1741688251495361, "logps/rejected": -1.359588384628296, "loss": 2.1736, "rewards/accuracies": 0.5625, "rewards/chosen": -1.1741688251495361, "rewards/margins": 0.18541957437992096, "rewards/rejected": -1.359588384628296, "step": 5 }, { "epoch": 0.02093692750588851, "grad_norm": 16.7263304813632, "learning_rate": 1.0416666666666667e-07, "logits/chosen": -0.5214197039604187, "logits/rejected": -0.49839648604393005, "logps/chosen": -1.1591269969940186, "logps/rejected": -1.2628146409988403, "loss": 2.1404, "rewards/accuracies": 0.5, "rewards/chosen": -1.1591269969940186, "rewards/margins": 0.10368760675191879, "rewards/rejected": -1.2628146409988403, "step": 10 }, { "epoch": 0.031405391258832765, "grad_norm": 31.370650570864015, "learning_rate": 1.5624999999999999e-07, "logits/chosen": -0.4619904160499573, "logits/rejected": -0.4506860673427582, "logps/chosen": -1.107530951499939, "logps/rejected": -1.362442970275879, "loss": 2.1079, "rewards/accuracies": 0.65625, "rewards/chosen": -1.107530951499939, "rewards/margins": 0.25491195917129517, "rewards/rejected": -1.362442970275879, "step": 15 }, { "epoch": 0.04187385501177702, "grad_norm": 39.77312899360463, "learning_rate": 2.0833333333333333e-07, "logits/chosen": -0.44072794914245605, "logits/rejected": -0.45300206542015076, "logps/chosen": -1.1610082387924194, "logps/rejected": -1.263232707977295, "loss": 2.1672, "rewards/accuracies": 0.5625, "rewards/chosen": -1.1610082387924194, "rewards/margins": 0.10222446918487549, "rewards/rejected": -1.263232707977295, "step": 20 }, { "epoch": 0.05234231876472128, "grad_norm": 13.297272362399367, "learning_rate": 2.604166666666667e-07, "logits/chosen": -0.5022198557853699, "logits/rejected": -0.47837525606155396, "logps/chosen": -1.1782314777374268, "logps/rejected": -1.241213083267212, "loss": 2.1428, "rewards/accuracies": 0.518750011920929, "rewards/chosen": -1.1782314777374268, "rewards/margins": 0.06298139691352844, "rewards/rejected": -1.241213083267212, "step": 25 }, { "epoch": 0.06281078251766553, "grad_norm": 37.00312129408131, "learning_rate": 3.1249999999999997e-07, "logits/chosen": -0.4893794059753418, "logits/rejected": -0.4941066801548004, "logps/chosen": -1.1618492603302002, "logps/rejected": -1.2827363014221191, "loss": 2.1704, "rewards/accuracies": 0.5062500238418579, "rewards/chosen": -1.1618492603302002, "rewards/margins": 0.12088712304830551, "rewards/rejected": -1.2827363014221191, "step": 30 }, { "epoch": 0.07327924627060979, "grad_norm": 19.474111043847227, "learning_rate": 3.645833333333333e-07, "logits/chosen": -0.4861086308956146, "logits/rejected": -0.44397059082984924, "logps/chosen": -1.146590232849121, "logps/rejected": -1.2768195867538452, "loss": 2.1372, "rewards/accuracies": 0.550000011920929, "rewards/chosen": -1.146590232849121, "rewards/margins": 0.13022944331169128, "rewards/rejected": -1.2768195867538452, "step": 35 }, { "epoch": 0.08374771002355404, "grad_norm": 27.757551759863073, "learning_rate": 4.1666666666666667e-07, "logits/chosen": -0.49622225761413574, "logits/rejected": -0.502428412437439, "logps/chosen": -1.0711848735809326, "logps/rejected": -1.3877671957015991, "loss": 2.0943, "rewards/accuracies": 0.643750011920929, "rewards/chosen": -1.0711848735809326, "rewards/margins": 0.31658223271369934, "rewards/rejected": -1.3877671957015991, "step": 40 }, { "epoch": 0.0942161737764983, "grad_norm": 29.522164186661577, "learning_rate": 4.6874999999999996e-07, "logits/chosen": -0.4612571597099304, "logits/rejected": -0.473367303609848, "logps/chosen": -1.0810363292694092, "logps/rejected": -1.2934114933013916, "loss": 2.1151, "rewards/accuracies": 0.59375, "rewards/chosen": -1.0810363292694092, "rewards/margins": 0.21237528324127197, "rewards/rejected": -1.2934114933013916, "step": 45 }, { "epoch": 0.10468463752944256, "grad_norm": 29.341931035274772, "learning_rate": 4.999731868769026e-07, "logits/chosen": -0.48391613364219666, "logits/rejected": -0.5020588040351868, "logps/chosen": -1.1583433151245117, "logps/rejected": -1.4378485679626465, "loss": 2.1398, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -1.1583433151245117, "rewards/margins": 0.2795052230358124, "rewards/rejected": -1.4378485679626465, "step": 50 }, { "epoch": 0.11515310128238682, "grad_norm": 123.21015696544787, "learning_rate": 4.996716052911017e-07, "logits/chosen": -0.4529895782470703, "logits/rejected": -0.44476404786109924, "logps/chosen": -1.1183184385299683, "logps/rejected": -1.3346388339996338, "loss": 2.1114, "rewards/accuracies": 0.5562499761581421, "rewards/chosen": -1.1183184385299683, "rewards/margins": 0.21632030606269836, "rewards/rejected": -1.3346388339996338, "step": 55 }, { "epoch": 0.12562156503533106, "grad_norm": 21.079325990888805, "learning_rate": 4.990353313429303e-07, "logits/chosen": -0.4931353032588959, "logits/rejected": -0.49371495842933655, "logps/chosen": -1.1979756355285645, "logps/rejected": -1.3100030422210693, "loss": 2.096, "rewards/accuracies": 0.5687500238418579, "rewards/chosen": -1.1979756355285645, "rewards/margins": 0.11202754080295563, "rewards/rejected": -1.3100030422210693, "step": 60 }, { "epoch": 0.1360900287882753, "grad_norm": 18.626119398460727, "learning_rate": 4.980652179769217e-07, "logits/chosen": -0.45450514554977417, "logits/rejected": -0.4266931414604187, "logps/chosen": -1.0912764072418213, "logps/rejected": -1.5846312046051025, "loss": 2.0791, "rewards/accuracies": 0.6875, "rewards/chosen": -1.0912764072418213, "rewards/margins": 0.4933546483516693, "rewards/rejected": -1.5846312046051025, "step": 65 }, { "epoch": 0.14655849254121958, "grad_norm": 52.35457634472626, "learning_rate": 4.967625656594781e-07, "logits/chosen": -0.3869231343269348, "logits/rejected": -0.3839837908744812, "logps/chosen": -1.165590524673462, "logps/rejected": -1.4081659317016602, "loss": 2.0592, "rewards/accuracies": 0.5625, "rewards/chosen": -1.165590524673462, "rewards/margins": 0.2425754815340042, "rewards/rejected": -1.4081659317016602, "step": 70 }, { "epoch": 0.15702695629416383, "grad_norm": 45.522892507761505, "learning_rate": 4.951291206355559e-07, "logits/chosen": -0.38980668783187866, "logits/rejected": -0.3762575685977936, "logps/chosen": -1.1448724269866943, "logps/rejected": -1.4742355346679688, "loss": 2.0746, "rewards/accuracies": 0.574999988079071, "rewards/chosen": -1.1448724269866943, "rewards/margins": 0.3293631970882416, "rewards/rejected": -1.4742355346679688, "step": 75 }, { "epoch": 0.16749542004710807, "grad_norm": 31.87949007278887, "learning_rate": 4.93167072587771e-07, "logits/chosen": -0.40593957901000977, "logits/rejected": -0.3936491012573242, "logps/chosen": -1.1574352979660034, "logps/rejected": -1.5279510021209717, "loss": 2.1051, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -1.1574352979660034, "rewards/margins": 0.3705156445503235, "rewards/rejected": -1.5279510021209717, "step": 80 }, { "epoch": 0.17796388380005235, "grad_norm": 21.704707598130327, "learning_rate": 4.908790517010636e-07, "logits/chosen": -0.4470198154449463, "logits/rejected": -0.4167211055755615, "logps/chosen": -1.1367292404174805, "logps/rejected": -1.450998306274414, "loss": 2.1069, "rewards/accuracies": 0.625, "rewards/chosen": -1.1367292404174805, "rewards/margins": 0.3142690062522888, "rewards/rejected": -1.450998306274414, "step": 85 }, { "epoch": 0.1884323475529966, "grad_norm": 8.653417747444738, "learning_rate": 4.882681251368548e-07, "logits/chosen": -0.4369986951351166, "logits/rejected": -0.4109964370727539, "logps/chosen": -1.1568020582199097, "logps/rejected": -1.4440120458602905, "loss": 2.0549, "rewards/accuracies": 0.668749988079071, "rewards/chosen": -1.1568020582199097, "rewards/margins": 0.2872098982334137, "rewards/rejected": -1.4440120458602905, "step": 90 }, { "epoch": 0.19890081130594087, "grad_norm": 32.43410976766361, "learning_rate": 4.853377929214243e-07, "logits/chosen": -0.4856724739074707, "logits/rejected": -0.409660667181015, "logps/chosen": -1.1660270690917969, "logps/rejected": -1.429152488708496, "loss": 2.0952, "rewards/accuracies": 0.5562499761581421, "rewards/chosen": -1.1660270690917969, "rewards/margins": 0.26312538981437683, "rewards/rejected": -1.429152488708496, "step": 95 }, { "epoch": 0.2093692750588851, "grad_norm": 20.24510226455269, "learning_rate": 4.820919832540181e-07, "logits/chosen": -0.43944352865219116, "logits/rejected": -0.4186181128025055, "logps/chosen": -1.1845619678497314, "logps/rejected": -1.6824769973754883, "loss": 2.0785, "rewards/accuracies": 0.5375000238418579, "rewards/chosen": -1.1845619678497314, "rewards/margins": 0.4979149401187897, "rewards/rejected": -1.6824769973754883, "step": 100 }, { "epoch": 0.21983773881182936, "grad_norm": 21.93099294182866, "learning_rate": 4.785350472409791e-07, "logits/chosen": -0.4231189787387848, "logits/rejected": -0.41250792145729065, "logps/chosen": -1.103410005569458, "logps/rejected": -1.5199437141418457, "loss": 2.0523, "rewards/accuracies": 0.6312500238418579, "rewards/chosen": -1.103410005569458, "rewards/margins": 0.4165337085723877, "rewards/rejected": -1.5199437141418457, "step": 105 }, { "epoch": 0.23030620256477363, "grad_norm": 192.6716268565829, "learning_rate": 4.7467175306295647e-07, "logits/chosen": -0.4392550587654114, "logits/rejected": -0.4161394238471985, "logps/chosen": -1.1035048961639404, "logps/rejected": -1.5209665298461914, "loss": 2.0465, "rewards/accuracies": 0.625, "rewards/chosen": -1.1035048961639404, "rewards/margins": 0.41746172308921814, "rewards/rejected": -1.5209665298461914, "step": 110 }, { "epoch": 0.24077466631771788, "grad_norm": 40.84875259149394, "learning_rate": 4.70507279583015e-07, "logits/chosen": -0.336051881313324, "logits/rejected": -0.3062174916267395, "logps/chosen": -1.0912038087844849, "logps/rejected": -1.6076161861419678, "loss": 2.0318, "rewards/accuracies": 0.65625, "rewards/chosen": -1.0912038087844849, "rewards/margins": 0.5164125561714172, "rewards/rejected": -1.6076161861419678, "step": 115 }, { "epoch": 0.2512431300706621, "grad_norm": 1121.915313367022, "learning_rate": 4.6604720940421207e-07, "logits/chosen": -0.31270137429237366, "logits/rejected": -0.29621389508247375, "logps/chosen": -1.258256435394287, "logps/rejected": -1.5046377182006836, "loss": 2.1046, "rewards/accuracies": 0.5687500238418579, "rewards/chosen": -1.258256435394287, "rewards/margins": 0.24638131260871887, "rewards/rejected": -1.5046377182006836, "step": 120 }, { "epoch": 0.26171159382360637, "grad_norm": 20.937334753207477, "learning_rate": 4.612975213859487e-07, "logits/chosen": -0.32645902037620544, "logits/rejected": -0.2999876141548157, "logps/chosen": -1.124294400215149, "logps/rejected": -1.4352140426635742, "loss": 2.0543, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -1.124294400215149, "rewards/margins": 0.3109196722507477, "rewards/rejected": -1.4352140426635742, "step": 125 }, { "epoch": 0.2721800575765506, "grad_norm": 123.08016970705063, "learning_rate": 4.5626458262912735e-07, "logits/chosen": -0.2738082706928253, "logits/rejected": -0.26782605051994324, "logps/chosen": -1.09318208694458, "logps/rejected": -1.3113747835159302, "loss": 2.08, "rewards/accuracies": 0.5687500238418579, "rewards/chosen": -1.09318208694458, "rewards/margins": 0.2181926667690277, "rewards/rejected": -1.3113747835159302, "step": 130 }, { "epoch": 0.2826485213294949, "grad_norm": 38.90891382468301, "learning_rate": 4.5095513994085974e-07, "logits/chosen": -0.33431217074394226, "logits/rejected": -0.25280579924583435, "logps/chosen": -1.1433501243591309, "logps/rejected": -1.7623112201690674, "loss": 2.0285, "rewards/accuracies": 0.737500011920929, "rewards/chosen": -1.1433501243591309, "rewards/margins": 0.618960976600647, "rewards/rejected": -1.7623112201690674, "step": 135 }, { "epoch": 0.29311698508243916, "grad_norm": 90.38260744944395, "learning_rate": 4.453763107901675e-07, "logits/chosen": -0.2669193148612976, "logits/rejected": -0.20890673995018005, "logps/chosen": -1.1904810667037964, "logps/rejected": -1.6921430826187134, "loss": 2.0201, "rewards/accuracies": 0.65625, "rewards/chosen": -1.1904810667037964, "rewards/margins": 0.5016618967056274, "rewards/rejected": -1.6921430826187134, "step": 140 }, { "epoch": 0.3035854488353834, "grad_norm": 23.623204151291763, "learning_rate": 4.395355737667985e-07, "logits/chosen": -0.3123199939727783, "logits/rejected": -0.29313474893569946, "logps/chosen": -1.1307309865951538, "logps/rejected": -1.4845192432403564, "loss": 2.0527, "rewards/accuracies": 0.643750011920929, "rewards/chosen": -1.1307309865951538, "rewards/margins": 0.3537880778312683, "rewards/rejected": -1.4845192432403564, "step": 145 }, { "epoch": 0.31405391258832765, "grad_norm": 20.156387548307503, "learning_rate": 4.3344075855595097e-07, "logits/chosen": -0.28625836968421936, "logits/rejected": -0.2564564049243927, "logps/chosen": -1.0314805507659912, "logps/rejected": -1.4487305879592896, "loss": 2.0217, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -1.0314805507659912, "rewards/margins": 0.41724997758865356, "rewards/rejected": -1.4487305879592896, "step": 150 }, { "epoch": 0.3245223763412719, "grad_norm": 51.259884987761694, "learning_rate": 4.271000354423425e-07, "logits/chosen": -0.3215218186378479, "logits/rejected": -0.26840001344680786, "logps/chosen": -1.142395257949829, "logps/rejected": -1.4793251752853394, "loss": 2.013, "rewards/accuracies": 0.5874999761581421, "rewards/chosen": -1.142395257949829, "rewards/margins": 0.33692988753318787, "rewards/rejected": -1.4793251752853394, "step": 155 }, { "epoch": 0.33499084009421615, "grad_norm": 36.42093295445508, "learning_rate": 4.2052190435769554e-07, "logits/chosen": -0.24544882774353027, "logits/rejected": -0.19806651771068573, "logps/chosen": -1.2047197818756104, "logps/rejected": -1.631598711013794, "loss": 2.0508, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.2047197818756104, "rewards/margins": 0.42687878012657166, "rewards/rejected": -1.631598711013794, "step": 160 }, { "epoch": 0.34545930384716045, "grad_norm": 39.07058204343973, "learning_rate": 4.137151834863213e-07, "logits/chosen": -0.3026384711265564, "logits/rejected": -0.25133180618286133, "logps/chosen": -1.1269896030426025, "logps/rejected": -1.6380598545074463, "loss": 2.0307, "rewards/accuracies": 0.606249988079071, "rewards/chosen": -1.1269896030426025, "rewards/margins": 0.5110700130462646, "rewards/rejected": -1.6380598545074463, "step": 165 }, { "epoch": 0.3559277676001047, "grad_norm": 30.659300325934307, "learning_rate": 4.0668899744407567e-07, "logits/chosen": -0.3210826516151428, "logits/rejected": -0.24679234623908997, "logps/chosen": -1.2077548503875732, "logps/rejected": -1.6277275085449219, "loss": 2.0405, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -1.2077548503875732, "rewards/margins": 0.4199727475643158, "rewards/rejected": -1.6277275085449219, "step": 170 }, { "epoch": 0.36639623135304894, "grad_norm": 29.84733882043046, "learning_rate": 3.994527650465352e-07, "logits/chosen": -0.3458993434906006, "logits/rejected": -0.25427910685539246, "logps/chosen": -1.1780593395233154, "logps/rejected": -1.5799932479858398, "loss": 2.0373, "rewards/accuracies": 0.5687500238418579, "rewards/chosen": -1.1780593395233154, "rewards/margins": 0.4019339084625244, "rewards/rejected": -1.5799932479858398, "step": 175 }, { "epoch": 0.3768646951059932, "grad_norm": 18.975102811277186, "learning_rate": 3.920161866827889e-07, "logits/chosen": -0.3150302767753601, "logits/rejected": -0.28204405307769775, "logps/chosen": -1.1404414176940918, "logps/rejected": -1.519669532775879, "loss": 2.0334, "rewards/accuracies": 0.625, "rewards/chosen": -1.1404414176940918, "rewards/margins": 0.3792281746864319, "rewards/rejected": -1.519669532775879, "step": 180 }, { "epoch": 0.38733315885893743, "grad_norm": 14.475409524552186, "learning_rate": 3.8438923131177237e-07, "logits/chosen": -0.282858669757843, "logits/rejected": -0.19747613370418549, "logps/chosen": -1.165867805480957, "logps/rejected": -1.813596487045288, "loss": 1.965, "rewards/accuracies": 0.625, "rewards/chosen": -1.165867805480957, "rewards/margins": 0.6477286219596863, "rewards/rejected": -1.813596487045288, "step": 185 }, { "epoch": 0.39780162261188173, "grad_norm": 59.63064263707996, "learning_rate": 3.765821230985757e-07, "logits/chosen": -0.30526572465896606, "logits/rejected": -0.23906362056732178, "logps/chosen": -1.2474441528320312, "logps/rejected": -1.6622447967529297, "loss": 2.107, "rewards/accuracies": 0.59375, "rewards/chosen": -1.2474441528320312, "rewards/margins": 0.4148005545139313, "rewards/rejected": -1.6622447967529297, "step": 190 }, { "epoch": 0.408270086364826, "grad_norm": 23.989688460552507, "learning_rate": 3.6860532770864005e-07, "logits/chosen": -0.26398172974586487, "logits/rejected": -0.21283094584941864, "logps/chosen": -1.2256332635879517, "logps/rejected": -1.5658466815948486, "loss": 2.0384, "rewards/accuracies": 0.6187499761581421, "rewards/chosen": -1.2256332635879517, "rewards/margins": 0.3402135968208313, "rewards/rejected": -1.5658466815948486, "step": 195 }, { "epoch": 0.4187385501177702, "grad_norm": 18.678426330461928, "learning_rate": 3.604695382782159e-07, "logits/chosen": -0.29794207215309143, "logits/rejected": -0.26024094223976135, "logps/chosen": -1.197186827659607, "logps/rejected": -1.5625463724136353, "loss": 2.0478, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -1.197186827659607, "rewards/margins": 0.36535948514938354, "rewards/rejected": -1.5625463724136353, "step": 200 }, { "epoch": 0.42920701387071447, "grad_norm": 18.836564429743568, "learning_rate": 3.5218566107988867e-07, "logits/chosen": -0.21744203567504883, "logits/rejected": -0.16317346692085266, "logps/chosen": -1.038894772529602, "logps/rejected": -1.5853629112243652, "loss": 1.9918, "rewards/accuracies": 0.6187499761581421, "rewards/chosen": -1.038894772529602, "rewards/margins": 0.5464681386947632, "rewards/rejected": -1.5853629112243652, "step": 205 }, { "epoch": 0.4396754776236587, "grad_norm": 72.20345838571406, "learning_rate": 3.4376480090239047e-07, "logits/chosen": -0.2059970200061798, "logits/rejected": -0.1603209674358368, "logps/chosen": -1.2739497423171997, "logps/rejected": -1.5597654581069946, "loss": 2.0481, "rewards/accuracies": 0.59375, "rewards/chosen": -1.2739497423171997, "rewards/margins": 0.28581586480140686, "rewards/rejected": -1.5597654581069946, "step": 210 }, { "epoch": 0.45014394137660296, "grad_norm": 27.301032186734012, "learning_rate": 3.3521824616429284e-07, "logits/chosen": -0.25597435235977173, "logits/rejected": -0.20625348389148712, "logps/chosen": -1.148979663848877, "logps/rejected": -1.677459478378296, "loss": 2.0204, "rewards/accuracies": 0.7437499761581421, "rewards/chosen": -1.148979663848877, "rewards/margins": 0.5284797549247742, "rewards/rejected": -1.677459478378296, "step": 215 }, { "epoch": 0.46061240512954726, "grad_norm": 29.629980256927464, "learning_rate": 3.265574537815398e-07, "logits/chosen": -0.3048572540283203, "logits/rejected": -0.1881858855485916, "logps/chosen": -1.165297269821167, "logps/rejected": -1.6496442556381226, "loss": 1.9967, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -1.165297269821167, "rewards/margins": 0.4843469262123108, "rewards/rejected": -1.6496442556381226, "step": 220 }, { "epoch": 0.4710808688824915, "grad_norm": 37.80136795892406, "learning_rate": 3.1779403380910425e-07, "logits/chosen": -0.2867579162120819, "logits/rejected": -0.19062136113643646, "logps/chosen": -1.084784746170044, "logps/rejected": -1.6818145513534546, "loss": 1.9613, "rewards/accuracies": 0.637499988079071, "rewards/chosen": -1.084784746170044, "rewards/margins": 0.5970298647880554, "rewards/rejected": -1.6818145513534546, "step": 225 }, { "epoch": 0.48154933263543576, "grad_norm": 35.96385036849735, "learning_rate": 3.0893973387735683e-07, "logits/chosen": -0.3329482972621918, "logits/rejected": -0.2059972733259201, "logps/chosen": -1.1524263620376587, "logps/rejected": -1.833388328552246, "loss": 1.9819, "rewards/accuracies": 0.643750011920929, "rewards/chosen": -1.1524263620376587, "rewards/margins": 0.6809619665145874, "rewards/rejected": -1.833388328552246, "step": 230 }, { "epoch": 0.49201779638838, "grad_norm": 22.085252727016336, "learning_rate": 3.000064234440111e-07, "logits/chosen": -0.24676939845085144, "logits/rejected": -0.17916962504386902, "logps/chosen": -1.1332640647888184, "logps/rejected": -1.6203981637954712, "loss": 2.0097, "rewards/accuracies": 0.6812499761581421, "rewards/chosen": -1.1332640647888184, "rewards/margins": 0.4871342182159424, "rewards/rejected": -1.6203981637954712, "step": 235 }, { "epoch": 0.5024862601413242, "grad_norm": 59.80174206160742, "learning_rate": 2.910060778827554e-07, "logits/chosen": -0.27282705903053284, "logits/rejected": -0.2174268662929535, "logps/chosen": -1.1373220682144165, "logps/rejected": -1.6652929782867432, "loss": 2.0531, "rewards/accuracies": 0.606249988079071, "rewards/chosen": -1.1373220682144165, "rewards/margins": 0.5279708504676819, "rewards/rejected": -1.6652929782867432, "step": 240 }, { "epoch": 0.5129547238942685, "grad_norm": 28.05175376412517, "learning_rate": 2.8195076242990116e-07, "logits/chosen": -0.3008952736854553, "logits/rejected": -0.22454960644245148, "logps/chosen": -1.229134202003479, "logps/rejected": -1.6271793842315674, "loss": 2.0109, "rewards/accuracies": 0.6187499761581421, "rewards/chosen": -1.229134202003479, "rewards/margins": 0.3980451822280884, "rewards/rejected": -1.6271793842315674, "step": 245 }, { "epoch": 0.5234231876472127, "grad_norm": 55.23162384622342, "learning_rate": 2.7285261601056697e-07, "logits/chosen": -0.23525886237621307, "logits/rejected": -0.1734374463558197, "logps/chosen": -1.1406811475753784, "logps/rejected": -1.7381279468536377, "loss": 2.0332, "rewards/accuracies": 0.6812499761581421, "rewards/chosen": -1.1406811475753784, "rewards/margins": 0.5974463224411011, "rewards/rejected": -1.7381279468536377, "step": 250 }, { "epoch": 0.533891651400157, "grad_norm": 42.787865372213936, "learning_rate": 2.6372383496608186e-07, "logits/chosen": -0.19053250551223755, "logits/rejected": -0.0688900500535965, "logps/chosen": -1.2063381671905518, "logps/rejected": -1.7581781148910522, "loss": 1.9736, "rewards/accuracies": 0.65625, "rewards/chosen": -1.2063381671905518, "rewards/margins": 0.5518399477005005, "rewards/rejected": -1.7581781148910522, "step": 255 }, { "epoch": 0.5443601151531012, "grad_norm": 66.05775717329243, "learning_rate": 2.5457665670441937e-07, "logits/chosen": -0.17143914103507996, "logits/rejected": -0.0884476974606514, "logps/chosen": -1.1227834224700928, "logps/rejected": -1.7763004302978516, "loss": 2.0199, "rewards/accuracies": 0.668749988079071, "rewards/chosen": -1.1227834224700928, "rewards/margins": 0.6535168886184692, "rewards/rejected": -1.7763004302978516, "step": 260 }, { "epoch": 0.5548285789060455, "grad_norm": 24.089434105436485, "learning_rate": 2.454233432955807e-07, "logits/chosen": -0.17798355221748352, "logits/rejected": -0.11273761838674545, "logps/chosen": -1.1982738971710205, "logps/rejected": -1.5514423847198486, "loss": 1.9891, "rewards/accuracies": 0.6312500238418579, "rewards/chosen": -1.1982738971710205, "rewards/margins": 0.35316845774650574, "rewards/rejected": -1.5514423847198486, "step": 265 }, { "epoch": 0.5652970426589898, "grad_norm": 38.85091217421057, "learning_rate": 2.3627616503391812e-07, "logits/chosen": -0.15943729877471924, "logits/rejected": -0.07409695535898209, "logps/chosen": -1.2726773023605347, "logps/rejected": -1.6484864950180054, "loss": 2.028, "rewards/accuracies": 0.6875, "rewards/chosen": -1.2726773023605347, "rewards/margins": 0.3758092522621155, "rewards/rejected": -1.6484864950180054, "step": 270 }, { "epoch": 0.575765506411934, "grad_norm": 16.063870661090185, "learning_rate": 2.2714738398943308e-07, "logits/chosen": -0.15996511280536652, "logits/rejected": -0.01324576698243618, "logps/chosen": -1.1747169494628906, "logps/rejected": -1.8413015604019165, "loss": 1.9966, "rewards/accuracies": 0.6875, "rewards/chosen": -1.1747169494628906, "rewards/margins": 0.6665846705436707, "rewards/rejected": -1.8413015604019165, "step": 275 }, { "epoch": 0.5862339701648783, "grad_norm": 83.57376579636319, "learning_rate": 2.1804923757009882e-07, "logits/chosen": -0.16753597557544708, "logits/rejected": -0.05278133228421211, "logps/chosen": -1.0996477603912354, "logps/rejected": -1.5048593282699585, "loss": 1.9938, "rewards/accuracies": 0.65625, "rewards/chosen": -1.0996477603912354, "rewards/margins": 0.40521159768104553, "rewards/rejected": -1.5048593282699585, "step": 280 }, { "epoch": 0.5967024339178225, "grad_norm": 33.49859165217226, "learning_rate": 2.089939221172446e-07, "logits/chosen": -0.10943803936243057, "logits/rejected": -0.07321613281965256, "logps/chosen": -1.190861463546753, "logps/rejected": -1.7705156803131104, "loss": 1.9823, "rewards/accuracies": 0.6625000238418579, "rewards/chosen": -1.190861463546753, "rewards/margins": 0.5796541571617126, "rewards/rejected": -1.7705156803131104, "step": 285 }, { "epoch": 0.6071708976707668, "grad_norm": 30.87334049891428, "learning_rate": 1.9999357655598891e-07, "logits/chosen": -0.1590811163187027, "logits/rejected": -0.07361742109060287, "logps/chosen": -1.181041955947876, "logps/rejected": -1.800238013267517, "loss": 1.9789, "rewards/accuracies": 0.6812499761581421, "rewards/chosen": -1.181041955947876, "rewards/margins": 0.6191961765289307, "rewards/rejected": -1.800238013267517, "step": 290 }, { "epoch": 0.6176393614237111, "grad_norm": 23.02159874411763, "learning_rate": 1.9106026612264315e-07, "logits/chosen": -0.16086161136627197, "logits/rejected": -0.08389837294816971, "logps/chosen": -1.208166480064392, "logps/rejected": -1.826172113418579, "loss": 2.0188, "rewards/accuracies": 0.637499988079071, "rewards/chosen": -1.208166480064392, "rewards/margins": 0.6180055737495422, "rewards/rejected": -1.826172113418579, "step": 295 }, { "epoch": 0.6281078251766553, "grad_norm": 31.936289365137483, "learning_rate": 1.8220596619089573e-07, "logits/chosen": -0.21206030249595642, "logits/rejected": -0.15204045176506042, "logps/chosen": -1.2225364446640015, "logps/rejected": -1.6840988397598267, "loss": 2.0085, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.2225364446640015, "rewards/margins": 0.46156245470046997, "rewards/rejected": -1.6840988397598267, "step": 300 }, { "epoch": 0.6385762889295996, "grad_norm": 40.76245078380065, "learning_rate": 1.7344254621846017e-07, "logits/chosen": -0.11568818241357803, "logits/rejected": -0.054085873067379, "logps/chosen": -1.2138392925262451, "logps/rejected": -1.8302034139633179, "loss": 1.9638, "rewards/accuracies": 0.6312500238418579, "rewards/chosen": -1.2138392925262451, "rewards/margins": 0.6163640022277832, "rewards/rejected": -1.8302034139633179, "step": 305 }, { "epoch": 0.6490447526825438, "grad_norm": 49.19939562684821, "learning_rate": 1.647817538357072e-07, "logits/chosen": -0.12115196883678436, "logits/rejected": -0.1311618536710739, "logps/chosen": -1.0947141647338867, "logps/rejected": -1.5839706659317017, "loss": 1.9698, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.0947141647338867, "rewards/margins": 0.48925653100013733, "rewards/rejected": -1.5839706659317017, "step": 310 }, { "epoch": 0.6595132164354881, "grad_norm": 27.360907272103443, "learning_rate": 1.562351990976095e-07, "logits/chosen": -0.1562352329492569, "logits/rejected": -0.04307347536087036, "logps/chosen": -1.0912151336669922, "logps/rejected": -1.8711601495742798, "loss": 1.9701, "rewards/accuracies": 0.6937500238418579, "rewards/chosen": -1.0912151336669922, "rewards/margins": 0.7799449563026428, "rewards/rejected": -1.8711601495742798, "step": 315 }, { "epoch": 0.6699816801884323, "grad_norm": 37.06696857881054, "learning_rate": 1.478143389201113e-07, "logits/chosen": -0.11155574023723602, "logits/rejected": -0.07031629979610443, "logps/chosen": -1.225824236869812, "logps/rejected": -1.715296745300293, "loss": 1.9507, "rewards/accuracies": 0.625, "rewards/chosen": -1.225824236869812, "rewards/margins": 0.48947247862815857, "rewards/rejected": -1.715296745300293, "step": 320 }, { "epoch": 0.6804501439413766, "grad_norm": 47.67894323163762, "learning_rate": 1.3953046172178413e-07, "logits/chosen": -0.14251194894313812, "logits/rejected": -0.046834878623485565, "logps/chosen": -1.1644965410232544, "logps/rejected": -1.8360859155654907, "loss": 1.9483, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.1644965410232544, "rewards/margins": 0.6715894341468811, "rewards/rejected": -1.8360859155654907, "step": 325 }, { "epoch": 0.6909186076943209, "grad_norm": 15.072577205919428, "learning_rate": 1.3139467229135998e-07, "logits/chosen": -0.11516846716403961, "logits/rejected": -0.03220799192786217, "logps/chosen": -1.144539713859558, "logps/rejected": -1.6052381992340088, "loss": 1.9932, "rewards/accuracies": 0.5874999761581421, "rewards/chosen": -1.144539713859558, "rewards/margins": 0.46069830656051636, "rewards/rejected": -1.6052381992340088, "step": 330 }, { "epoch": 0.7013870714472651, "grad_norm": 34.23220245935474, "learning_rate": 1.2341787690142435e-07, "logits/chosen": -0.1980099380016327, "logits/rejected": -0.10996635258197784, "logps/chosen": -1.110797643661499, "logps/rejected": -1.5681132078170776, "loss": 1.9936, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.110797643661499, "rewards/margins": 0.4573155343532562, "rewards/rejected": -1.5681132078170776, "step": 335 }, { "epoch": 0.7118555352002094, "grad_norm": 28.671371339841, "learning_rate": 1.1561076868822755e-07, "logits/chosen": -0.18400368094444275, "logits/rejected": -0.09802932292222977, "logps/chosen": -1.275649070739746, "logps/rejected": -1.726248025894165, "loss": 2.023, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.275649070739746, "rewards/margins": 0.45059889554977417, "rewards/rejected": -1.726248025894165, "step": 340 }, { "epoch": 0.7223239989531536, "grad_norm": 33.94958056858452, "learning_rate": 1.0798381331721107e-07, "logits/chosen": -0.20783087611198425, "logits/rejected": -0.0951487272977829, "logps/chosen": -1.164470911026001, "logps/rejected": -1.7271054983139038, "loss": 2.001, "rewards/accuracies": 0.6187499761581421, "rewards/chosen": -1.164470911026001, "rewards/margins": 0.5626345872879028, "rewards/rejected": -1.7271054983139038, "step": 345 }, { "epoch": 0.7327924627060979, "grad_norm": 21.893648949832478, "learning_rate": 1.0054723495346482e-07, "logits/chosen": -0.07672537118196487, "logits/rejected": -0.03784745931625366, "logps/chosen": -1.1689984798431396, "logps/rejected": -1.6384124755859375, "loss": 2.0337, "rewards/accuracies": 0.606249988079071, "rewards/chosen": -1.1689984798431396, "rewards/margins": 0.469414085149765, "rewards/rejected": -1.6384124755859375, "step": 350 }, { "epoch": 0.7432609264590422, "grad_norm": 32.78382225954148, "learning_rate": 9.331100255592436e-08, "logits/chosen": -0.07636372745037079, "logits/rejected": 0.007526999805122614, "logps/chosen": -1.0614162683486938, "logps/rejected": -1.7131534814834595, "loss": 1.9457, "rewards/accuracies": 0.71875, "rewards/chosen": -1.0614162683486938, "rewards/margins": 0.6517372727394104, "rewards/rejected": -1.7131534814834595, "step": 355 }, { "epoch": 0.7537293902119864, "grad_norm": 18.581615763356663, "learning_rate": 8.628481651367875e-08, "logits/chosen": -0.165523499250412, "logits/rejected": -0.06427756696939468, "logps/chosen": -1.054750680923462, "logps/rejected": -1.56973397731781, "loss": 1.9192, "rewards/accuracies": 0.6625000238418579, "rewards/chosen": -1.054750680923462, "rewards/margins": 0.5149833559989929, "rewards/rejected": -1.56973397731781, "step": 360 }, { "epoch": 0.7641978539649307, "grad_norm": 33.79193132717166, "learning_rate": 7.947809564230445e-08, "logits/chosen": -0.15916807949543, "logits/rejected": -0.04110134392976761, "logps/chosen": -1.2637778520584106, "logps/rejected": -1.8154550790786743, "loss": 2.0161, "rewards/accuracies": 0.6312500238418579, "rewards/chosen": -1.2637778520584106, "rewards/margins": 0.5516773462295532, "rewards/rejected": -1.8154550790786743, "step": 365 }, { "epoch": 0.7746663177178749, "grad_norm": 30.598444258335533, "learning_rate": 7.289996455765748e-08, "logits/chosen": -0.09070632606744766, "logits/rejected": -0.016309842467308044, "logps/chosen": -1.233588695526123, "logps/rejected": -1.8405319452285767, "loss": 1.9721, "rewards/accuracies": 0.6875, "rewards/chosen": -1.233588695526123, "rewards/margins": 0.6069431900978088, "rewards/rejected": -1.8405319452285767, "step": 370 }, { "epoch": 0.7851347814708192, "grad_norm": 47.76183473164956, "learning_rate": 6.655924144404906e-08, "logits/chosen": -0.12095741182565689, "logits/rejected": -0.15326176583766937, "logps/chosen": -1.2032678127288818, "logps/rejected": -1.698957085609436, "loss": 1.9989, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -1.2032678127288818, "rewards/margins": 0.49568939208984375, "rewards/rejected": -1.698957085609436, "step": 375 }, { "epoch": 0.7956032452237635, "grad_norm": 26.82763374923621, "learning_rate": 6.046442623320145e-08, "logits/chosen": -0.16168245673179626, "logits/rejected": -0.10053972899913788, "logps/chosen": -1.2546281814575195, "logps/rejected": -1.6799224615097046, "loss": 1.9554, "rewards/accuracies": 0.581250011920929, "rewards/chosen": -1.2546281814575195, "rewards/margins": 0.42529433965682983, "rewards/rejected": -1.6799224615097046, "step": 380 }, { "epoch": 0.8060717089767077, "grad_norm": 22.887488172337097, "learning_rate": 5.4623689209832484e-08, "logits/chosen": -0.1701394021511078, "logits/rejected": -0.10942293703556061, "logps/chosen": -1.1635555028915405, "logps/rejected": -1.6771953105926514, "loss": 1.9821, "rewards/accuracies": 0.59375, "rewards/chosen": -1.1635555028915405, "rewards/margins": 0.5136396884918213, "rewards/rejected": -1.6771953105926514, "step": 385 }, { "epoch": 0.816540172729652, "grad_norm": 17.93102818717875, "learning_rate": 4.904486005914027e-08, "logits/chosen": -0.16732355952262878, "logits/rejected": -0.0791902020573616, "logps/chosen": -1.1596777439117432, "logps/rejected": -1.8085283041000366, "loss": 1.9712, "rewards/accuracies": 0.699999988079071, "rewards/chosen": -1.1596777439117432, "rewards/margins": 0.6488505005836487, "rewards/rejected": -1.8085283041000366, "step": 390 }, { "epoch": 0.8270086364825961, "grad_norm": 45.13149823297093, "learning_rate": 4.373541737087263e-08, "logits/chosen": -0.17222358286380768, "logits/rejected": -0.02697933092713356, "logps/chosen": -1.18060302734375, "logps/rejected": -1.7580945491790771, "loss": 1.9717, "rewards/accuracies": 0.6812499761581421, "rewards/chosen": -1.18060302734375, "rewards/margins": 0.5774916410446167, "rewards/rejected": -1.7580945491790771, "step": 395 }, { "epoch": 0.8374771002355405, "grad_norm": 17.576421623633806, "learning_rate": 3.8702478614051345e-08, "logits/chosen": -0.10352808237075806, "logits/rejected": -0.04908682033419609, "logps/chosen": -1.10788893699646, "logps/rejected": -1.7277934551239014, "loss": 1.9503, "rewards/accuracies": 0.643750011920929, "rewards/chosen": -1.10788893699646, "rewards/margins": 0.619904637336731, "rewards/rejected": -1.7277934551239014, "step": 400 }, { "epoch": 0.8479455639884846, "grad_norm": 26.16217732840323, "learning_rate": 3.3952790595787986e-08, "logits/chosen": -0.197185680270195, "logits/rejected": -0.12642063200473785, "logps/chosen": -1.2650173902511597, "logps/rejected": -1.8639274835586548, "loss": 2.0076, "rewards/accuracies": 0.65625, "rewards/chosen": -1.2650173902511597, "rewards/margins": 0.5989099740982056, "rewards/rejected": -1.8639274835586548, "step": 405 }, { "epoch": 0.8584140277414289, "grad_norm": 24.116020361120558, "learning_rate": 2.9492720416985e-08, "logits/chosen": -0.1570913940668106, "logits/rejected": -0.08443228900432587, "logps/chosen": -1.1111314296722412, "logps/rejected": -1.5838778018951416, "loss": 1.9708, "rewards/accuracies": 0.606249988079071, "rewards/chosen": -1.1111314296722412, "rewards/margins": 0.47274643182754517, "rewards/rejected": -1.5838778018951416, "step": 410 }, { "epoch": 0.8688824914943732, "grad_norm": 23.795655447251885, "learning_rate": 2.5328246937043525e-08, "logits/chosen": -0.1644868105649948, "logits/rejected": -0.05437507480382919, "logps/chosen": -1.1470693349838257, "logps/rejected": -1.7496206760406494, "loss": 1.9309, "rewards/accuracies": 0.65625, "rewards/chosen": -1.1470693349838257, "rewards/margins": 0.6025511622428894, "rewards/rejected": -1.7496206760406494, "step": 415 }, { "epoch": 0.8793509552473174, "grad_norm": 17.858809164573977, "learning_rate": 2.1464952759020856e-08, "logits/chosen": -0.15058764815330505, "logits/rejected": -0.0765046551823616, "logps/chosen": -1.1509023904800415, "logps/rejected": -1.7408424615859985, "loss": 1.9941, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -1.1509023904800415, "rewards/margins": 0.5899400115013123, "rewards/rejected": -1.7408424615859985, "step": 420 }, { "epoch": 0.8898194190002617, "grad_norm": 100.72607171931102, "learning_rate": 1.7908016745981856e-08, "logits/chosen": -0.11002065241336823, "logits/rejected": -0.048098642379045486, "logps/chosen": -1.2740198373794556, "logps/rejected": -1.741000771522522, "loss": 1.9691, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -1.2740198373794556, "rewards/margins": 0.46698078513145447, "rewards/rejected": -1.741000771522522, "step": 425 }, { "epoch": 0.9002878827532059, "grad_norm": 22.399845767434137, "learning_rate": 1.4662207078575684e-08, "logits/chosen": -0.14560401439666748, "logits/rejected": -0.08761245012283325, "logps/chosen": -1.198464035987854, "logps/rejected": -1.7667903900146484, "loss": 1.9534, "rewards/accuracies": 0.625, "rewards/chosen": -1.198464035987854, "rewards/margins": 0.568326473236084, "rewards/rejected": -1.7667903900146484, "step": 430 }, { "epoch": 0.9107563465061502, "grad_norm": 26.789977249333305, "learning_rate": 1.1731874863145142e-08, "logits/chosen": -0.1761171966791153, "logits/rejected": -0.11556446552276611, "logps/chosen": -1.1400909423828125, "logps/rejected": -1.924628496170044, "loss": 1.9424, "rewards/accuracies": 0.6499999761581421, "rewards/chosen": -1.1400909423828125, "rewards/margins": 0.7845373153686523, "rewards/rejected": -1.924628496170044, "step": 435 }, { "epoch": 0.9212248102590945, "grad_norm": 29.975236886779925, "learning_rate": 9.12094829893642e-09, "logits/chosen": -0.023710379377007484, "logits/rejected": -0.008227316662669182, "logps/chosen": -1.2032636404037476, "logps/rejected": -1.6965421438217163, "loss": 2.0197, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -1.2032636404037476, "rewards/margins": 0.4932785630226135, "rewards/rejected": -1.6965421438217163, "step": 440 }, { "epoch": 0.9316932740120387, "grad_norm": 44.53364841982202, "learning_rate": 6.832927412229017e-09, "logits/chosen": -0.1587805449962616, "logits/rejected": -0.09741976112127304, "logps/chosen": -1.186359167098999, "logps/rejected": -1.6756283044815063, "loss": 2.0179, "rewards/accuracies": 0.625, "rewards/chosen": -1.186359167098999, "rewards/margins": 0.4892690181732178, "rewards/rejected": -1.6756283044815063, "step": 445 }, { "epoch": 0.942161737764983, "grad_norm": 37.66309860666725, "learning_rate": 4.8708793644441086e-09, "logits/chosen": -0.07072281092405319, "logits/rejected": -0.03446631878614426, "logps/chosen": -1.2110041379928589, "logps/rejected": -1.6703050136566162, "loss": 1.9597, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -1.2110041379928589, "rewards/margins": 0.4593009054660797, "rewards/rejected": -1.6703050136566162, "step": 450 }, { "epoch": 0.9526302015179272, "grad_norm": 59.24673385214879, "learning_rate": 3.2374343405217884e-09, "logits/chosen": -0.12624625861644745, "logits/rejected": -0.06765826046466827, "logps/chosen": -1.3134384155273438, "logps/rejected": -1.6258878707885742, "loss": 2.0031, "rewards/accuracies": 0.6187499761581421, "rewards/chosen": -1.3134384155273438, "rewards/margins": 0.31244954466819763, "rewards/rejected": -1.6258878707885742, "step": 455 }, { "epoch": 0.9630986652708715, "grad_norm": 31.809388609630044, "learning_rate": 1.9347820230782295e-09, "logits/chosen": -0.09453988820314407, "logits/rejected": -0.033892810344696045, "logps/chosen": -1.093733549118042, "logps/rejected": -1.6776697635650635, "loss": 1.9354, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -1.093733549118042, "rewards/margins": 0.5839362144470215, "rewards/rejected": -1.6776697635650635, "step": 460 }, { "epoch": 0.9735671290238157, "grad_norm": 48.638368475214634, "learning_rate": 9.64668657069706e-10, "logits/chosen": -0.19474738836288452, "logits/rejected": -0.08652818948030472, "logps/chosen": -1.2114652395248413, "logps/rejected": -1.7437559366226196, "loss": 1.9666, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -1.2114652395248413, "rewards/margins": 0.5322908163070679, "rewards/rejected": -1.7437559366226196, "step": 465 }, { "epoch": 0.98403559277676, "grad_norm": 20.81325024609551, "learning_rate": 3.2839470889836627e-10, "logits/chosen": -0.17209358513355255, "logits/rejected": -0.08596277236938477, "logps/chosen": -1.142816185951233, "logps/rejected": -1.610129952430725, "loss": 2.0171, "rewards/accuracies": 0.637499988079071, "rewards/chosen": -1.142816185951233, "rewards/margins": 0.46731385588645935, "rewards/rejected": -1.610129952430725, "step": 470 }, { "epoch": 0.9945040565297043, "grad_norm": 22.432046053055338, "learning_rate": 2.6813123097352287e-11, "logits/chosen": -0.20779597759246826, "logits/rejected": -0.07040806114673615, "logps/chosen": -1.2208012342453003, "logps/rejected": -1.6825767755508423, "loss": 1.9899, "rewards/accuracies": 0.581250011920929, "rewards/chosen": -1.2208012342453003, "rewards/margins": 0.46177545189857483, "rewards/rejected": -1.6825767755508423, "step": 475 }, { "epoch": 0.998691442030882, "step": 477, "total_flos": 0.0, "train_loss": 0.0, "train_runtime": 4.2983, "train_samples_per_second": 14223.022, "train_steps_per_second": 110.974 } ], "logging_steps": 5, "max_steps": 477, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 1000000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }