{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.887991927346115, "eval_steps": 50, "global_step": 1100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.008072653884964682, "grad_norm": 0.04926518723368645, "learning_rate": 4.999451708687114e-06, "logits/chosen": 14.542106628417969, "logits/rejected": 14.864250183105469, "logps/chosen": -0.2809702754020691, "logps/rejected": -0.3013763725757599, "loss": 0.9221, "rewards/accuracies": 0.42500001192092896, "rewards/chosen": -0.421455442905426, "rewards/margins": 0.03060910664498806, "rewards/rejected": -0.452064573764801, "step": 10 }, { "epoch": 0.016145307769929364, "grad_norm": 0.056249432265758514, "learning_rate": 4.997807075247147e-06, "logits/chosen": 14.614748001098633, "logits/rejected": 15.259109497070312, "logps/chosen": -0.2828002870082855, "logps/rejected": -0.3477819561958313, "loss": 0.916, "rewards/accuracies": 0.550000011920929, "rewards/chosen": -0.4242004454135895, "rewards/margins": 0.09747247397899628, "rewards/rejected": -0.5216729044914246, "step": 20 }, { "epoch": 0.024217961654894045, "grad_norm": 0.0664869099855423, "learning_rate": 4.9950668210706795e-06, "logits/chosen": 14.391204833984375, "logits/rejected": 14.82734203338623, "logps/chosen": -0.287629634141922, "logps/rejected": -0.3329126834869385, "loss": 0.914, "rewards/accuracies": 0.48750001192092896, "rewards/chosen": -0.4314444661140442, "rewards/margins": 0.06792456656694412, "rewards/rejected": -0.4993689954280853, "step": 30 }, { "epoch": 0.03229061553985873, "grad_norm": 0.055584829300642014, "learning_rate": 4.9912321481237616e-06, "logits/chosen": 13.93278980255127, "logits/rejected": 14.886846542358398, "logps/chosen": -0.28155946731567383, "logps/rejected": -0.3678051829338074, "loss": 0.9273, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -0.42233920097351074, "rewards/margins": 0.12936851382255554, "rewards/rejected": -0.5517078042030334, "step": 40 }, { "epoch": 0.04036326942482341, "grad_norm": 0.07997103035449982, "learning_rate": 4.986304738420684e-06, "logits/chosen": 14.172693252563477, "logits/rejected": 14.742494583129883, "logps/chosen": -0.28225988149642944, "logps/rejected": -0.34329456090927124, "loss": 0.9081, "rewards/accuracies": 0.48750001192092896, "rewards/chosen": -0.42338982224464417, "rewards/margins": 0.0915520042181015, "rewards/rejected": -0.5149418115615845, "step": 50 }, { "epoch": 0.04036326942482341, "eval_logits/chosen": 14.200166702270508, "eval_logits/rejected": 14.817726135253906, "eval_logps/chosen": -0.2625390887260437, "eval_logps/rejected": -0.3458769917488098, "eval_loss": 0.9080610275268555, "eval_rewards/accuracies": 0.5544554591178894, "eval_rewards/chosen": -0.39380866289138794, "eval_rewards/margins": 0.1250067949295044, "eval_rewards/rejected": -0.5188154578208923, "eval_runtime": 29.8098, "eval_samples_per_second": 26.87, "eval_steps_per_second": 3.388, "step": 50 }, { "epoch": 0.04843592330978809, "grad_norm": 0.06322095543146133, "learning_rate": 4.980286753286196e-06, "logits/chosen": 14.008010864257812, "logits/rejected": 14.975939750671387, "logps/chosen": -0.2697351574897766, "logps/rejected": -0.3445274233818054, "loss": 0.9195, "rewards/accuracies": 0.5, "rewards/chosen": -0.40460270643234253, "rewards/margins": 0.1121884360909462, "rewards/rejected": -0.5167912244796753, "step": 60 }, { "epoch": 0.056508577194752774, "grad_norm": 0.12975476682186127, "learning_rate": 4.973180832407471e-06, "logits/chosen": 14.213134765625, "logits/rejected": 14.889978408813477, "logps/chosen": -0.2839818596839905, "logps/rejected": -0.3382417559623718, "loss": 0.9241, "rewards/accuracies": 0.5375000238418579, "rewards/chosen": -0.4259727895259857, "rewards/margins": 0.08138985931873322, "rewards/rejected": -0.5073626637458801, "step": 70 }, { "epoch": 0.06458123107971746, "grad_norm": 0.08352109789848328, "learning_rate": 4.964990092676263e-06, "logits/chosen": 14.395146369934082, "logits/rejected": 14.900177001953125, "logps/chosen": -0.27067264914512634, "logps/rejected": -0.3516673743724823, "loss": 0.9251, "rewards/accuracies": 0.637499988079071, "rewards/chosen": -0.4060089588165283, "rewards/margins": 0.12149210274219513, "rewards/rejected": -0.5275009870529175, "step": 80 }, { "epoch": 0.07265388496468214, "grad_norm": 0.0793827548623085, "learning_rate": 4.9557181268217225e-06, "logits/chosen": 13.905145645141602, "logits/rejected": 14.897878646850586, "logps/chosen": -0.26814645528793335, "logps/rejected": -0.38061466813087463, "loss": 0.9055, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -0.4022197127342224, "rewards/margins": 0.16870227456092834, "rewards/rejected": -0.5709219574928284, "step": 90 }, { "epoch": 0.08072653884964683, "grad_norm": 0.09212008118629456, "learning_rate": 4.9453690018345144e-06, "logits/chosen": 14.11627197265625, "logits/rejected": 14.571977615356445, "logps/chosen": -0.3130624294281006, "logps/rejected": -0.35073375701904297, "loss": 0.911, "rewards/accuracies": 0.48750001192092896, "rewards/chosen": -0.46959367394447327, "rewards/margins": 0.05650699883699417, "rewards/rejected": -0.5261006951332092, "step": 100 }, { "epoch": 0.08072653884964683, "eval_logits/chosen": 13.845876693725586, "eval_logits/rejected": 14.490789413452148, "eval_logps/chosen": -0.2531408667564392, "eval_logps/rejected": -0.3464036285877228, "eval_loss": 0.8986235857009888, "eval_rewards/accuracies": 0.5544554591178894, "eval_rewards/chosen": -0.3797112703323364, "eval_rewards/margins": 0.13989417254924774, "eval_rewards/rejected": -0.5196054577827454, "eval_runtime": 29.0914, "eval_samples_per_second": 27.534, "eval_steps_per_second": 3.472, "step": 100 }, { "epoch": 0.08879919273461151, "grad_norm": 0.08554862439632416, "learning_rate": 4.933947257182901e-06, "logits/chosen": 13.416229248046875, "logits/rejected": 14.582674026489258, "logps/chosen": -0.2523443102836609, "logps/rejected": -0.38751405477523804, "loss": 0.8993, "rewards/accuracies": 0.6875, "rewards/chosen": -0.3785164952278137, "rewards/margins": 0.20275457203388214, "rewards/rejected": -0.5812710523605347, "step": 110 }, { "epoch": 0.09687184661957618, "grad_norm": 0.29209578037261963, "learning_rate": 4.921457902821578e-06, "logits/chosen": 13.874654769897461, "logits/rejected": 14.423624992370605, "logps/chosen": -0.27300480008125305, "logps/rejected": -0.3526575267314911, "loss": 0.8942, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -0.40950721502304077, "rewards/margins": 0.11947910487651825, "rewards/rejected": -0.5289863348007202, "step": 120 }, { "epoch": 0.10494450050454086, "grad_norm": 0.1028478816151619, "learning_rate": 4.907906416994146e-06, "logits/chosen": 14.030723571777344, "logits/rejected": 14.711235046386719, "logps/chosen": -0.27410784363746643, "logps/rejected": -0.3665519058704376, "loss": 0.8922, "rewards/accuracies": 0.550000011920929, "rewards/chosen": -0.41116175055503845, "rewards/margins": 0.1386660784482956, "rewards/rejected": -0.5498278737068176, "step": 130 }, { "epoch": 0.11301715438950555, "grad_norm": 0.08459590375423431, "learning_rate": 4.893298743830168e-06, "logits/chosen": 13.477182388305664, "logits/rejected": 13.919464111328125, "logps/chosen": -0.2495063841342926, "logps/rejected": -0.3594816029071808, "loss": 0.8957, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -0.3742595613002777, "rewards/margins": 0.16496284306049347, "rewards/rejected": -0.5392224192619324, "step": 140 }, { "epoch": 0.12108980827447023, "grad_norm": 0.12467797845602036, "learning_rate": 4.8776412907378845e-06, "logits/chosen": 12.801656723022461, "logits/rejected": 13.564155578613281, "logps/chosen": -0.2779986262321472, "logps/rejected": -0.33566445112228394, "loss": 0.892, "rewards/accuracies": 0.4375, "rewards/chosen": -0.4169979691505432, "rewards/margins": 0.08649872243404388, "rewards/rejected": -0.5034967064857483, "step": 150 }, { "epoch": 0.12108980827447023, "eval_logits/chosen": 12.169118881225586, "eval_logits/rejected": 13.019683837890625, "eval_logps/chosen": -0.25547870993614197, "eval_logps/rejected": -0.37277930974960327, "eval_loss": 0.8789658546447754, "eval_rewards/accuracies": 0.5742574334144592, "eval_rewards/chosen": -0.38321802020072937, "eval_rewards/margins": 0.17595094442367554, "eval_rewards/rejected": -0.5591689944267273, "eval_runtime": 29.0961, "eval_samples_per_second": 27.529, "eval_steps_per_second": 3.471, "step": 150 }, { "epoch": 0.12916246215943492, "grad_norm": 0.1712370663881302, "learning_rate": 4.860940925593703e-06, "logits/chosen": 11.972528457641602, "logits/rejected": 12.684088706970215, "logps/chosen": -0.28340521454811096, "logps/rejected": -0.3864063024520874, "loss": 0.8789, "rewards/accuracies": 0.574999988079071, "rewards/chosen": -0.42510780692100525, "rewards/margins": 0.15450166165828705, "rewards/rejected": -0.5796095132827759, "step": 160 }, { "epoch": 0.13723511604439959, "grad_norm": 0.21750673651695251, "learning_rate": 4.84320497372973e-06, "logits/chosen": 11.93881607055664, "logits/rejected": 12.458230972290039, "logps/chosen": -0.274599552154541, "logps/rejected": -0.3882916271686554, "loss": 0.8857, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -0.4118993282318115, "rewards/margins": 0.1705380380153656, "rewards/rejected": -0.5824374556541443, "step": 170 }, { "epoch": 0.14530776992936428, "grad_norm": 0.16000741720199585, "learning_rate": 4.824441214720629e-06, "logits/chosen": 9.631464958190918, "logits/rejected": 10.772969245910645, "logps/chosen": -0.287865549325943, "logps/rejected": -0.4817379415035248, "loss": 0.8699, "rewards/accuracies": 0.6625000238418579, "rewards/chosen": -0.4317983090877533, "rewards/margins": 0.29080861806869507, "rewards/rejected": -0.722606897354126, "step": 180 }, { "epoch": 0.15338042381432895, "grad_norm": 0.46293890476226807, "learning_rate": 4.804657878971252e-06, "logits/chosen": 9.081937789916992, "logits/rejected": 10.024572372436523, "logps/chosen": -0.2941994071006775, "logps/rejected": -0.4772109389305115, "loss": 0.8565, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -0.44129911065101624, "rewards/margins": 0.2745172679424286, "rewards/rejected": -0.7158163785934448, "step": 190 }, { "epoch": 0.16145307769929365, "grad_norm": 0.24658434092998505, "learning_rate": 4.783863644106502e-06, "logits/chosen": 7.767125606536865, "logits/rejected": 8.009145736694336, "logps/chosen": -0.32877305150032043, "logps/rejected": -0.47733697295188904, "loss": 0.8482, "rewards/accuracies": 0.5874999761581421, "rewards/chosen": -0.49315959215164185, "rewards/margins": 0.2228458821773529, "rewards/rejected": -0.7160054445266724, "step": 200 }, { "epoch": 0.16145307769929365, "eval_logits/chosen": 6.716187953948975, "eval_logits/rejected": 7.248146057128906, "eval_logps/chosen": -0.30652713775634766, "eval_logps/rejected": -0.5056277513504028, "eval_loss": 0.8243693113327026, "eval_rewards/accuracies": 0.603960394859314, "eval_rewards/chosen": -0.4597907066345215, "eval_rewards/margins": 0.2986510097980499, "eval_rewards/rejected": -0.7584417462348938, "eval_runtime": 29.0989, "eval_samples_per_second": 27.527, "eval_steps_per_second": 3.471, "step": 200 }, { "epoch": 0.16952573158425832, "grad_norm": 0.40796443819999695, "learning_rate": 4.762067631165049e-06, "logits/chosen": 5.932644844055176, "logits/rejected": 6.521953582763672, "logps/chosen": -0.32976508140563965, "logps/rejected": -0.5628186464309692, "loss": 0.7988, "rewards/accuracies": 0.612500011920929, "rewards/chosen": -0.4946475923061371, "rewards/margins": 0.3495803475379944, "rewards/rejected": -0.8442279696464539, "step": 210 }, { "epoch": 0.17759838546922302, "grad_norm": 0.42468318343162537, "learning_rate": 4.7392794005985324e-06, "logits/chosen": 6.069305896759033, "logits/rejected": 5.8950395584106445, "logps/chosen": -0.37205421924591064, "logps/rejected": -0.6190425753593445, "loss": 0.796, "rewards/accuracies": 0.6000000238418579, "rewards/chosen": -0.5580812692642212, "rewards/margins": 0.3704826235771179, "rewards/rejected": -0.9285639524459839, "step": 220 }, { "epoch": 0.1856710393541877, "grad_norm": 0.4138280153274536, "learning_rate": 4.715508948078037e-06, "logits/chosen": 3.5271706581115723, "logits/rejected": 3.363534927368164, "logps/chosen": -0.41677650809288025, "logps/rejected": -0.7121980786323547, "loss": 0.7457, "rewards/accuracies": 0.637499988079071, "rewards/chosen": -0.6251648664474487, "rewards/margins": 0.4431324005126953, "rewards/rejected": -1.0682971477508545, "step": 230 }, { "epoch": 0.19374369323915236, "grad_norm": 1.4327284097671509, "learning_rate": 4.690766700109659e-06, "logits/chosen": 2.296924114227295, "logits/rejected": 1.6135867834091187, "logps/chosen": -0.506227970123291, "logps/rejected": -0.8080593943595886, "loss": 0.7453, "rewards/accuracies": 0.550000011920929, "rewards/chosen": -0.7593418955802917, "rewards/margins": 0.4527471661567688, "rewards/rejected": -1.21208918094635, "step": 240 }, { "epoch": 0.20181634712411706, "grad_norm": 0.4413074553012848, "learning_rate": 4.665063509461098e-06, "logits/chosen": 1.4450469017028809, "logits/rejected": 0.40727120637893677, "logps/chosen": -0.4749869406223297, "logps/rejected": -0.9967275857925415, "loss": 0.704, "rewards/accuracies": 0.6625000238418579, "rewards/chosen": -0.712480366230011, "rewards/margins": 0.7826108932495117, "rewards/rejected": -1.4950913190841675, "step": 250 }, { "epoch": 0.20181634712411706, "eval_logits/chosen": 1.4614256620407104, "eval_logits/rejected": 0.6895493865013123, "eval_logps/chosen": -0.5038881301879883, "eval_logps/rejected": -0.973581850528717, "eval_loss": 0.6982013583183289, "eval_rewards/accuracies": 0.6138613820075989, "eval_rewards/chosen": -0.7558321952819824, "eval_rewards/margins": 0.7045406699180603, "eval_rewards/rejected": -1.460372805595398, "eval_runtime": 29.0993, "eval_samples_per_second": 27.526, "eval_steps_per_second": 3.471, "step": 250 }, { "epoch": 0.20988900100908173, "grad_norm": 0.734251856803894, "learning_rate": 4.638410650401267e-06, "logits/chosen": 2.4274039268493652, "logits/rejected": 1.498230218887329, "logps/chosen": -0.5549699068069458, "logps/rejected": -0.9348627328872681, "loss": 0.7149, "rewards/accuracies": 0.5249999761581421, "rewards/chosen": -0.8324548602104187, "rewards/margins": 0.5698392987251282, "rewards/rejected": -1.4022941589355469, "step": 260 }, { "epoch": 0.21796165489404642, "grad_norm": 0.4445085823535919, "learning_rate": 4.610819813755038e-06, "logits/chosen": 2.042858600616455, "logits/rejected": 1.1695036888122559, "logps/chosen": -0.5573975443840027, "logps/rejected": -1.15065598487854, "loss": 0.6811, "rewards/accuracies": 0.625, "rewards/chosen": -0.8360962867736816, "rewards/margins": 0.889887809753418, "rewards/rejected": -1.72598397731781, "step": 270 }, { "epoch": 0.2260343087790111, "grad_norm": 0.5875476598739624, "learning_rate": 4.582303101775249e-06, "logits/chosen": 1.3029518127441406, "logits/rejected": 0.3808741271495819, "logps/chosen": -0.6656169891357422, "logps/rejected": -1.5305824279785156, "loss": 0.637, "rewards/accuracies": 0.625, "rewards/chosen": -0.9984253644943237, "rewards/margins": 1.2974482774734497, "rewards/rejected": -2.2958736419677734, "step": 280 }, { "epoch": 0.2341069626639758, "grad_norm": 0.3814420998096466, "learning_rate": 4.55287302283426e-06, "logits/chosen": 2.0200212001800537, "logits/rejected": 1.2681838274002075, "logps/chosen": -0.6776013970375061, "logps/rejected": -1.3369777202606201, "loss": 0.63, "rewards/accuracies": 0.5874999761581421, "rewards/chosen": -1.016402006149292, "rewards/margins": 0.9890643358230591, "rewards/rejected": -2.0054664611816406, "step": 290 }, { "epoch": 0.24217961654894046, "grad_norm": 0.6442322731018066, "learning_rate": 4.522542485937369e-06, "logits/chosen": 1.6268196105957031, "logits/rejected": 0.4954712390899658, "logps/chosen": -0.8117648363113403, "logps/rejected": -1.7705228328704834, "loss": 0.6104, "rewards/accuracies": 0.7749999761581421, "rewards/chosen": -1.2176473140716553, "rewards/margins": 1.4381370544433594, "rewards/rejected": -2.6557843685150146, "step": 300 }, { "epoch": 0.24217961654894046, "eval_logits/chosen": 1.4003607034683228, "eval_logits/rejected": 0.46628010272979736, "eval_logps/chosen": -0.7612115740776062, "eval_logps/rejected": -1.6895866394042969, "eval_loss": 0.5781419277191162, "eval_rewards/accuracies": 0.6534653306007385, "eval_rewards/chosen": -1.141817331314087, "eval_rewards/margins": 1.392562747001648, "eval_rewards/rejected": -2.5343799591064453, "eval_runtime": 29.102, "eval_samples_per_second": 27.524, "eval_steps_per_second": 3.471, "step": 300 }, { "epoch": 0.25025227043390513, "grad_norm": 0.7889758944511414, "learning_rate": 4.491324795060491e-06, "logits/chosen": 1.1702089309692383, "logits/rejected": 0.354276180267334, "logps/chosen": -0.8091352581977844, "logps/rejected": -1.944819450378418, "loss": 0.5807, "rewards/accuracies": 0.6875, "rewards/chosen": -1.2137027978897095, "rewards/margins": 1.7035261392593384, "rewards/rejected": -2.9172286987304688, "step": 310 }, { "epoch": 0.25832492431886983, "grad_norm": 1.410145878791809, "learning_rate": 4.4592336433146e-06, "logits/chosen": 2.0981459617614746, "logits/rejected": 1.2289329767227173, "logps/chosen": -0.8151634931564331, "logps/rejected": -1.977259874343872, "loss": 0.5007, "rewards/accuracies": 0.6625000238418579, "rewards/chosen": -1.222745418548584, "rewards/margins": 1.7431443929672241, "rewards/rejected": -2.9658896923065186, "step": 320 }, { "epoch": 0.26639757820383453, "grad_norm": 0.5564689040184021, "learning_rate": 4.426283106939474e-06, "logits/chosen": 1.7921768426895142, "logits/rejected": 0.7705962061882019, "logps/chosen": -0.9244564771652222, "logps/rejected": -2.5274672508239746, "loss": 0.4757, "rewards/accuracies": 0.75, "rewards/chosen": -1.3866846561431885, "rewards/margins": 2.4045166969299316, "rewards/rejected": -3.791201114654541, "step": 330 }, { "epoch": 0.27447023208879917, "grad_norm": 0.7554243803024292, "learning_rate": 4.3924876391293915e-06, "logits/chosen": 1.947997808456421, "logits/rejected": 0.9661592245101929, "logps/chosen": -0.9154227375984192, "logps/rejected": -3.0766491889953613, "loss": 0.4581, "rewards/accuracies": 0.7749999761581421, "rewards/chosen": -1.3731342554092407, "rewards/margins": 3.24183988571167, "rewards/rejected": -4.614974021911621, "step": 340 }, { "epoch": 0.28254288597376387, "grad_norm": 2.936426877975464, "learning_rate": 4.357862063693486e-06, "logits/chosen": 1.2846364974975586, "logits/rejected": 0.9968118667602539, "logps/chosen": -1.0272481441497803, "logps/rejected": -2.6980624198913574, "loss": 0.4989, "rewards/accuracies": 0.7124999761581421, "rewards/chosen": -1.54087233543396, "rewards/margins": 2.506221294403076, "rewards/rejected": -4.047093868255615, "step": 350 }, { "epoch": 0.28254288597376387, "eval_logits/chosen": 1.841333031654358, "eval_logits/rejected": 1.1651691198349, "eval_logps/chosen": -0.9644113183021545, "eval_logps/rejected": -2.631535053253174, "eval_loss": 0.44154587388038635, "eval_rewards/accuracies": 0.6732673048973083, "eval_rewards/chosen": -1.4466170072555542, "eval_rewards/margins": 2.500684976577759, "eval_rewards/rejected": -3.9473025798797607, "eval_runtime": 29.1062, "eval_samples_per_second": 27.52, "eval_steps_per_second": 3.47, "step": 350 }, { "epoch": 0.29061553985872857, "grad_norm": 0.49919044971466064, "learning_rate": 4.322421568553529e-06, "logits/chosen": 1.9874728918075562, "logits/rejected": 1.464611291885376, "logps/chosen": -0.9963932037353516, "logps/rejected": -2.8804218769073486, "loss": 0.4373, "rewards/accuracies": 0.7124999761581421, "rewards/chosen": -1.4945898056030273, "rewards/margins": 2.8260433673858643, "rewards/rejected": -4.320633411407471, "step": 360 }, { "epoch": 0.29868819374369326, "grad_norm": 1.4991668462753296, "learning_rate": 4.286181699082008e-06, "logits/chosen": 2.5421957969665527, "logits/rejected": 1.941922903060913, "logps/chosen": -1.0214306116104126, "logps/rejected": -3.2893283367156982, "loss": 0.438, "rewards/accuracies": 0.7250000238418579, "rewards/chosen": -1.5321458578109741, "rewards/margins": 3.4018466472625732, "rewards/rejected": -4.933992862701416, "step": 370 }, { "epoch": 0.3067608476286579, "grad_norm": 1.933100938796997, "learning_rate": 4.249158351283414e-06, "logits/chosen": 1.80439031124115, "logits/rejected": 1.4421275854110718, "logps/chosen": -1.1270357370376587, "logps/rejected": -3.17500901222229, "loss": 0.4542, "rewards/accuracies": 0.7124999761581421, "rewards/chosen": -1.6905533075332642, "rewards/margins": 3.071959972381592, "rewards/rejected": -4.762513160705566, "step": 380 }, { "epoch": 0.3148335015136226, "grad_norm": 3.017254590988159, "learning_rate": 4.211367764821722e-06, "logits/chosen": 3.4090209007263184, "logits/rejected": 2.784639835357666, "logps/chosen": -1.1713608503341675, "logps/rejected": -3.326244831085205, "loss": 0.4816, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -1.757041335105896, "rewards/margins": 3.232325792312622, "rewards/rejected": -4.9893670082092285, "step": 390 }, { "epoch": 0.3229061553985873, "grad_norm": 0.5897337198257446, "learning_rate": 4.172826515897146e-06, "logits/chosen": 3.6071503162384033, "logits/rejected": 2.738395929336548, "logps/chosen": -1.133429765701294, "logps/rejected": -3.509474277496338, "loss": 0.4431, "rewards/accuracies": 0.7250000238418579, "rewards/chosen": -1.7001447677612305, "rewards/margins": 3.5640671253204346, "rewards/rejected": -5.264212131500244, "step": 400 }, { "epoch": 0.3229061553985873, "eval_logits/chosen": 2.893691062927246, "eval_logits/rejected": 2.259718656539917, "eval_logps/chosen": -1.0705492496490479, "eval_logps/rejected": -2.935904026031494, "eval_loss": 0.4156961143016815, "eval_rewards/accuracies": 0.6732673048973083, "eval_rewards/chosen": -1.6058237552642822, "eval_rewards/margins": 2.798032283782959, "eval_rewards/rejected": -4.403855800628662, "eval_runtime": 29.186, "eval_samples_per_second": 27.445, "eval_steps_per_second": 3.461, "step": 400 }, { "epoch": 0.33097880928355194, "grad_norm": 2.219709634780884, "learning_rate": 4.133551509975264e-06, "logits/chosen": 2.8940415382385254, "logits/rejected": 2.545257091522217, "logps/chosen": -1.1774877309799194, "logps/rejected": -3.4022269248962402, "loss": 0.427, "rewards/accuracies": 0.699999988079071, "rewards/chosen": -1.7662317752838135, "rewards/margins": 3.3371081352233887, "rewards/rejected": -5.1033406257629395, "step": 410 }, { "epoch": 0.33905146316851664, "grad_norm": 0.9891649484634399, "learning_rate": 4.093559974371725e-06, "logits/chosen": 3.180427074432373, "logits/rejected": 2.8354597091674805, "logps/chosen": -1.2984545230865479, "logps/rejected": -3.4927687644958496, "loss": 0.4384, "rewards/accuracies": 0.75, "rewards/chosen": -1.9476817846298218, "rewards/margins": 3.291471481323242, "rewards/rejected": -5.2391533851623535, "step": 420 }, { "epoch": 0.34712411705348134, "grad_norm": 0.4516288936138153, "learning_rate": 4.052869450695776e-06, "logits/chosen": 4.026360511779785, "logits/rejected": 3.3932456970214844, "logps/chosen": -1.1195746660232544, "logps/rejected": -3.4704582691192627, "loss": 0.3934, "rewards/accuracies": 0.6625000238418579, "rewards/chosen": -1.6793619394302368, "rewards/margins": 3.5263259410858154, "rewards/rejected": -5.205687522888184, "step": 430 }, { "epoch": 0.35519677093844604, "grad_norm": 0.5276560187339783, "learning_rate": 4.011497787155938e-06, "logits/chosen": 3.410902500152588, "logits/rejected": 2.906996965408325, "logps/chosen": -1.205904245376587, "logps/rejected": -3.7434115409851074, "loss": 0.4147, "rewards/accuracies": 0.737500011920929, "rewards/chosen": -1.8088566064834595, "rewards/margins": 3.806260585784912, "rewards/rejected": -5.615117073059082, "step": 440 }, { "epoch": 0.3632694248234107, "grad_norm": 0.4435446560382843, "learning_rate": 3.969463130731183e-06, "logits/chosen": 3.7616615295410156, "logits/rejected": 3.4183363914489746, "logps/chosen": -1.2180852890014648, "logps/rejected": -3.865541934967041, "loss": 0.4354, "rewards/accuracies": 0.675000011920929, "rewards/chosen": -1.8271278142929077, "rewards/margins": 3.9711856842041016, "rewards/rejected": -5.798312664031982, "step": 450 }, { "epoch": 0.3632694248234107, "eval_logits/chosen": 3.55381178855896, "eval_logits/rejected": 3.01943039894104, "eval_logps/chosen": -1.3239827156066895, "eval_logps/rejected": -3.6477084159851074, "eval_loss": 0.3905698359012604, "eval_rewards/accuracies": 0.7029703259468079, "eval_rewards/chosen": -1.9859741926193237, "eval_rewards/margins": 3.485589027404785, "eval_rewards/rejected": -5.471563816070557, "eval_runtime": 29.0958, "eval_samples_per_second": 27.53, "eval_steps_per_second": 3.471, "step": 450 }, { "epoch": 0.3713420787083754, "grad_norm": 0.8419796824455261, "learning_rate": 3.92678391921108e-06, "logits/chosen": 3.3496861457824707, "logits/rejected": 3.235544204711914, "logps/chosen": -1.4899990558624268, "logps/rejected": -4.507351398468018, "loss": 0.4023, "rewards/accuracies": 0.762499988079071, "rewards/chosen": -2.2349984645843506, "rewards/margins": 4.526029109954834, "rewards/rejected": -6.7610273361206055, "step": 460 }, { "epoch": 0.3794147325933401, "grad_norm": 1.2379052639007568, "learning_rate": 3.88347887310836e-06, "logits/chosen": 3.5375685691833496, "logits/rejected": 3.34586763381958, "logps/chosen": -1.5865886211395264, "logps/rejected": -4.887473106384277, "loss": 0.3979, "rewards/accuracies": 0.7875000238418579, "rewards/chosen": -2.379883289337158, "rewards/margins": 4.9513258934021, "rewards/rejected": -7.331210136413574, "step": 470 }, { "epoch": 0.3874873864783047, "grad_norm": 1.9969083070755005, "learning_rate": 3.839566987447492e-06, "logits/chosen": 3.3116583824157715, "logits/rejected": 2.9652373790740967, "logps/chosen": -2.0900139808654785, "logps/rejected": -4.921124458312988, "loss": 0.3356, "rewards/accuracies": 0.824999988079071, "rewards/chosen": -3.135021209716797, "rewards/margins": 4.246665000915527, "rewards/rejected": -7.381686210632324, "step": 480 }, { "epoch": 0.3955600403632694, "grad_norm": 2.3405466079711914, "learning_rate": 3.795067523432826e-06, "logits/chosen": 2.542064666748047, "logits/rejected": 2.503020763397217, "logps/chosen": -2.4674277305603027, "logps/rejected": -5.703408718109131, "loss": 0.3193, "rewards/accuracies": 0.875, "rewards/chosen": -3.701141834259033, "rewards/margins": 4.853971004486084, "rewards/rejected": -8.555112838745117, "step": 490 }, { "epoch": 0.4036326942482341, "grad_norm": 1.4798164367675781, "learning_rate": 3.7500000000000005e-06, "logits/chosen": 3.79949688911438, "logits/rejected": 3.7085328102111816, "logps/chosen": -2.6355605125427246, "logps/rejected": -5.005209922790527, "loss": 0.3103, "rewards/accuracies": 0.8125, "rewards/chosen": -3.953340530395508, "rewards/margins": 3.554474353790283, "rewards/rejected": -7.507814884185791, "step": 500 }, { "epoch": 0.4036326942482341, "eval_logits/chosen": 3.31948184967041, "eval_logits/rejected": 2.9104762077331543, "eval_logps/chosen": -2.606774091720581, "eval_logps/rejected": -5.2280120849609375, "eval_loss": 0.3228474259376526, "eval_rewards/accuracies": 0.9306930899620056, "eval_rewards/chosen": -3.9101614952087402, "eval_rewards/margins": 3.931856870651245, "eval_rewards/rejected": -7.8420186042785645, "eval_runtime": 29.0975, "eval_samples_per_second": 27.528, "eval_steps_per_second": 3.471, "step": 500 }, { "epoch": 0.4117053481331988, "grad_norm": 1.4098985195159912, "learning_rate": 3.7043841852542884e-06, "logits/chosen": 2.69830584526062, "logits/rejected": 2.7101688385009766, "logps/chosen": -2.7831828594207764, "logps/rejected": -5.963825225830078, "loss": 0.2884, "rewards/accuracies": 0.887499988079071, "rewards/chosen": -4.174774169921875, "rewards/margins": 4.7709641456604, "rewards/rejected": -8.945737838745117, "step": 510 }, { "epoch": 0.41977800201816345, "grad_norm": 3.457231283187866, "learning_rate": 3.658240087799655e-06, "logits/chosen": 3.335204601287842, "logits/rejected": 3.1935055255889893, "logps/chosen": -2.66495943069458, "logps/rejected": -5.2761640548706055, "loss": 0.3093, "rewards/accuracies": 0.875, "rewards/chosen": -3.997438907623291, "rewards/margins": 3.916806697845459, "rewards/rejected": -7.91424560546875, "step": 520 }, { "epoch": 0.42785065590312815, "grad_norm": 1.7744654417037964, "learning_rate": 3.611587947962319e-06, "logits/chosen": 2.725924015045166, "logits/rejected": 2.605053186416626, "logps/chosen": -2.5286645889282227, "logps/rejected": -5.774144172668457, "loss": 0.2972, "rewards/accuracies": 0.887499988079071, "rewards/chosen": -3.792996883392334, "rewards/margins": 4.868220329284668, "rewards/rejected": -8.661216735839844, "step": 530 }, { "epoch": 0.43592330978809285, "grad_norm": 2.863692283630371, "learning_rate": 3.564448228912682e-06, "logits/chosen": 2.304443120956421, "logits/rejected": 2.4836020469665527, "logps/chosen": -2.699148654937744, "logps/rejected": -6.340291500091553, "loss": 0.3136, "rewards/accuracies": 0.887499988079071, "rewards/chosen": -4.048723220825195, "rewards/margins": 5.461714267730713, "rewards/rejected": -9.510437965393066, "step": 540 }, { "epoch": 0.4439959636730575, "grad_norm": 4.6681623458862305, "learning_rate": 3.516841607689501e-06, "logits/chosen": 3.1207594871520996, "logits/rejected": 2.947653293609619, "logps/chosen": -3.0062317848205566, "logps/rejected": -6.26816463470459, "loss": 0.2822, "rewards/accuracies": 0.8999999761581421, "rewards/chosen": -4.509347438812256, "rewards/margins": 4.892899513244629, "rewards/rejected": -9.402246475219727, "step": 550 }, { "epoch": 0.4439959636730575, "eval_logits/chosen": 2.879721164703369, "eval_logits/rejected": 2.52551007270813, "eval_logps/chosen": -2.6468751430511475, "eval_logps/rejected": -5.636789321899414, "eval_loss": 0.2939932942390442, "eval_rewards/accuracies": 0.9306930899620056, "eval_rewards/chosen": -3.9703128337860107, "eval_rewards/margins": 4.4848713874816895, "eval_rewards/rejected": -8.455184936523438, "eval_runtime": 29.4025, "eval_samples_per_second": 27.243, "eval_steps_per_second": 3.435, "step": 550 }, { "epoch": 0.4520686175580222, "grad_norm": 1.9975569248199463, "learning_rate": 3.4687889661302577e-06, "logits/chosen": 2.5885415077209473, "logits/rejected": 2.5374302864074707, "logps/chosen": -2.886643171310425, "logps/rejected": -6.636415004730225, "loss": 0.2736, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -4.329964637756348, "rewards/margins": 5.624656677246094, "rewards/rejected": -9.954621315002441, "step": 560 }, { "epoch": 0.4601412714429869, "grad_norm": 3.3965606689453125, "learning_rate": 3.4203113817116955e-06, "logits/chosen": 2.0093374252319336, "logits/rejected": 2.1064200401306152, "logps/chosen": -3.251981735229492, "logps/rejected": -6.749331474304199, "loss": 0.2776, "rewards/accuracies": 0.875, "rewards/chosen": -4.877972602844238, "rewards/margins": 5.246025085449219, "rewards/rejected": -10.12399673461914, "step": 570 }, { "epoch": 0.4682139253279516, "grad_norm": 2.7079927921295166, "learning_rate": 3.3714301183045382e-06, "logits/chosen": 3.916841506958008, "logits/rejected": 3.524440288543701, "logps/chosen": -3.304469347000122, "logps/rejected": -7.1993279457092285, "loss": 0.2547, "rewards/accuracies": 0.9375, "rewards/chosen": -4.956704139709473, "rewards/margins": 5.842287540435791, "rewards/rejected": -10.798992156982422, "step": 580 }, { "epoch": 0.47628657921291623, "grad_norm": 7.504203796386719, "learning_rate": 3.3221666168464584e-06, "logits/chosen": 2.7350666522979736, "logits/rejected": 2.2590534687042236, "logps/chosen": -3.0954322814941406, "logps/rejected": -6.782465934753418, "loss": 0.2839, "rewards/accuracies": 0.887499988079071, "rewards/chosen": -4.643147945404053, "rewards/margins": 5.530551433563232, "rewards/rejected": -10.173700332641602, "step": 590 }, { "epoch": 0.4843592330978809, "grad_norm": 2.8209569454193115, "learning_rate": 3.272542485937369e-06, "logits/chosen": 3.7743606567382812, "logits/rejected": 3.2480530738830566, "logps/chosen": -3.196892261505127, "logps/rejected": -6.42517614364624, "loss": 0.3237, "rewards/accuracies": 0.875, "rewards/chosen": -4.7953386306762695, "rewards/margins": 4.8424248695373535, "rewards/rejected": -9.637763977050781, "step": 600 }, { "epoch": 0.4843592330978809, "eval_logits/chosen": 2.6344597339630127, "eval_logits/rejected": 2.46531081199646, "eval_logps/chosen": -2.9877941608428955, "eval_logps/rejected": -6.266376972198486, "eval_loss": 0.2827170193195343, "eval_rewards/accuracies": 0.9603960514068604, "eval_rewards/chosen": -4.481690406799316, "eval_rewards/margins": 4.917874336242676, "eval_rewards/rejected": -9.399564743041992, "eval_runtime": 29.0997, "eval_samples_per_second": 27.526, "eval_steps_per_second": 3.471, "step": 600 }, { "epoch": 0.4924318869828456, "grad_norm": 7.072656631469727, "learning_rate": 3.222579492361179e-06, "logits/chosen": 3.2776103019714355, "logits/rejected": 3.166206121444702, "logps/chosen": -3.193716049194336, "logps/rejected": -6.185642242431641, "loss": 0.2986, "rewards/accuracies": 0.8500000238418579, "rewards/chosen": -4.790574073791504, "rewards/margins": 4.487889289855957, "rewards/rejected": -9.278463363647461, "step": 610 }, { "epoch": 0.5005045408678103, "grad_norm": 2.7188704013824463, "learning_rate": 3.1722995515381644e-06, "logits/chosen": 3.189532518386841, "logits/rejected": 2.9885735511779785, "logps/chosen": -3.3461785316467285, "logps/rejected": -6.720322608947754, "loss": 0.2776, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.019268035888672, "rewards/margins": 5.061214447021484, "rewards/rejected": -10.080483436584473, "step": 620 }, { "epoch": 0.508577194752775, "grad_norm": 2.6990387439727783, "learning_rate": 3.121724717912138e-06, "logits/chosen": 3.7751071453094482, "logits/rejected": 3.406925916671753, "logps/chosen": -3.3688464164733887, "logps/rejected": -6.132623672485352, "loss": 0.2688, "rewards/accuracies": 0.8374999761581421, "rewards/chosen": -5.053269386291504, "rewards/margins": 4.145665168762207, "rewards/rejected": -9.198934555053711, "step": 630 }, { "epoch": 0.5166498486377397, "grad_norm": 1.964301347732544, "learning_rate": 3.0708771752766397e-06, "logits/chosen": 2.6602158546447754, "logits/rejected": 2.162348508834839, "logps/chosen": -2.884718894958496, "logps/rejected": -6.869751930236816, "loss": 0.2435, "rewards/accuracies": 0.9125000238418579, "rewards/chosen": -4.327078342437744, "rewards/margins": 5.9775495529174805, "rewards/rejected": -10.304628372192383, "step": 640 }, { "epoch": 0.5247225025227044, "grad_norm": 3.201704740524292, "learning_rate": 3.019779227044398e-06, "logits/chosen": 3.83849835395813, "logits/rejected": 3.5917961597442627, "logps/chosen": -3.2030792236328125, "logps/rejected": -6.711284637451172, "loss": 0.2865, "rewards/accuracies": 0.8999999761581421, "rewards/chosen": -4.804618835449219, "rewards/margins": 5.262308120727539, "rewards/rejected": -10.066926956176758, "step": 650 }, { "epoch": 0.5247225025227044, "eval_logits/chosen": 2.6539251804351807, "eval_logits/rejected": 2.3508856296539307, "eval_logps/chosen": -3.0629732608795166, "eval_logps/rejected": -6.429784297943115, "eval_loss": 0.25649499893188477, "eval_rewards/accuracies": 0.9702970385551453, "eval_rewards/chosen": -4.594459533691406, "eval_rewards/margins": 5.0502166748046875, "eval_rewards/rejected": -9.644676208496094, "eval_runtime": 29.1031, "eval_samples_per_second": 27.523, "eval_steps_per_second": 3.47, "step": 650 }, { "epoch": 0.5327951564076691, "grad_norm": 3.517667055130005, "learning_rate": 2.9684532864643123e-06, "logits/chosen": 2.804810047149658, "logits/rejected": 2.8151614665985107, "logps/chosen": -3.4792568683624268, "logps/rejected": -7.08618688583374, "loss": 0.243, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -5.2188849449157715, "rewards/margins": 5.41039514541626, "rewards/rejected": -10.629280090332031, "step": 660 }, { "epoch": 0.5408678102926338, "grad_norm": 8.366011619567871, "learning_rate": 2.9169218667902562e-06, "logits/chosen": 2.1079938411712646, "logits/rejected": 2.2689383029937744, "logps/chosen": -3.3151912689208984, "logps/rejected": -7.454249382019043, "loss": 0.266, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -4.972787380218506, "rewards/margins": 6.208587169647217, "rewards/rejected": -11.181374549865723, "step": 670 }, { "epoch": 0.5489404641775983, "grad_norm": 4.161479473114014, "learning_rate": 2.8652075714060296e-06, "logits/chosen": 2.1858832836151123, "logits/rejected": 2.1688101291656494, "logps/chosen": -3.3753318786621094, "logps/rejected": -6.9990692138671875, "loss": 0.2356, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.0629987716674805, "rewards/margins": 5.435605525970459, "rewards/rejected": -10.498603820800781, "step": 680 }, { "epoch": 0.557013118062563, "grad_norm": 2.156621217727661, "learning_rate": 2.813333083910761e-06, "logits/chosen": 2.5004096031188965, "logits/rejected": 2.289215087890625, "logps/chosen": -3.711970567703247, "logps/rejected": -7.685518741607666, "loss": 0.2286, "rewards/accuracies": 0.9375, "rewards/chosen": -5.56795597076416, "rewards/margins": 5.960321426391602, "rewards/rejected": -11.528276443481445, "step": 690 }, { "epoch": 0.5650857719475277, "grad_norm": 4.547532081604004, "learning_rate": 2.761321158169134e-06, "logits/chosen": 3.112332820892334, "logits/rejected": 2.685454845428467, "logps/chosen": -3.510646343231201, "logps/rejected": -7.8125505447387695, "loss": 0.2222, "rewards/accuracies": 0.9125000238418579, "rewards/chosen": -5.265969276428223, "rewards/margins": 6.452856540679932, "rewards/rejected": -11.718826293945312, "step": 700 }, { "epoch": 0.5650857719475277, "eval_logits/chosen": 2.7667236328125, "eval_logits/rejected": 2.4983408451080322, "eval_logps/chosen": -3.292154550552368, "eval_logps/rejected": -6.814427375793457, "eval_loss": 0.24892139434814453, "eval_rewards/accuracies": 0.9702970385551453, "eval_rewards/chosen": -4.938231468200684, "eval_rewards/margins": 5.2834086418151855, "eval_rewards/rejected": -10.221639633178711, "eval_runtime": 29.0862, "eval_samples_per_second": 27.539, "eval_steps_per_second": 3.472, "step": 700 }, { "epoch": 0.5731584258324924, "grad_norm": 3.3842532634735107, "learning_rate": 2.70919460833079e-06, "logits/chosen": 2.6863622665405273, "logits/rejected": 2.7381534576416016, "logps/chosen": -3.1926321983337402, "logps/rejected": -7.823538780212402, "loss": 0.2153, "rewards/accuracies": 0.9375, "rewards/chosen": -4.788949012756348, "rewards/margins": 6.946360111236572, "rewards/rejected": -11.735308647155762, "step": 710 }, { "epoch": 0.5812310797174571, "grad_norm": 2.0056378841400146, "learning_rate": 2.6569762988232838e-06, "logits/chosen": 2.938141345977783, "logits/rejected": 2.8052525520324707, "logps/chosen": -3.4339518547058105, "logps/rejected": -7.353487968444824, "loss": 0.2422, "rewards/accuracies": 0.875, "rewards/chosen": -5.150927543640137, "rewards/margins": 5.879303932189941, "rewards/rejected": -11.030232429504395, "step": 720 }, { "epoch": 0.5893037336024218, "grad_norm": 2.9281482696533203, "learning_rate": 2.604689134322999e-06, "logits/chosen": 3.5540454387664795, "logits/rejected": 3.6734695434570312, "logps/chosen": -3.0852925777435303, "logps/rejected": -6.869669437408447, "loss": 0.2694, "rewards/accuracies": 0.8500000238418579, "rewards/chosen": -4.627939224243164, "rewards/margins": 5.676565647125244, "rewards/rejected": -10.304505348205566, "step": 730 }, { "epoch": 0.5973763874873865, "grad_norm": 3.3841099739074707, "learning_rate": 2.5523560497083927e-06, "logits/chosen": 2.7171874046325684, "logits/rejected": 2.4911468029022217, "logps/chosen": -3.4943466186523438, "logps/rejected": -7.486982822418213, "loss": 0.2184, "rewards/accuracies": 0.9375, "rewards/chosen": -5.241519927978516, "rewards/margins": 5.988953113555908, "rewards/rejected": -11.230472564697266, "step": 740 }, { "epoch": 0.6054490413723511, "grad_norm": 2.6915197372436523, "learning_rate": 2.5e-06, "logits/chosen": 2.866816759109497, "logits/rejected": 2.84656023979187, "logps/chosen": -3.155074119567871, "logps/rejected": -8.019740104675293, "loss": 0.2364, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -4.732610702514648, "rewards/margins": 7.297000885009766, "rewards/rejected": -12.029611587524414, "step": 750 }, { "epoch": 0.6054490413723511, "eval_logits/chosen": 2.7122883796691895, "eval_logits/rejected": 2.4663355350494385, "eval_logps/chosen": -3.362666606903076, "eval_logps/rejected": -7.107527732849121, "eval_loss": 0.2374112904071808, "eval_rewards/accuracies": 0.9801980257034302, "eval_rewards/chosen": -5.043999671936035, "eval_rewards/margins": 5.617292881011963, "eval_rewards/rejected": -10.661293983459473, "eval_runtime": 29.0963, "eval_samples_per_second": 27.529, "eval_steps_per_second": 3.471, "step": 750 }, { "epoch": 0.6135216952573158, "grad_norm": 5.086276531219482, "learning_rate": 2.447643950291608e-06, "logits/chosen": 3.0712199211120605, "logits/rejected": 2.7986550331115723, "logps/chosen": -3.759767532348633, "logps/rejected": -8.20389175415039, "loss": 0.2467, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.639651298522949, "rewards/margins": 6.666187286376953, "rewards/rejected": -12.305837631225586, "step": 760 }, { "epoch": 0.6215943491422805, "grad_norm": 4.272627830505371, "learning_rate": 2.3953108656770018e-06, "logits/chosen": 2.4291179180145264, "logits/rejected": 2.247910737991333, "logps/chosen": -3.4369187355041504, "logps/rejected": -8.507123947143555, "loss": 0.2287, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -5.155378341674805, "rewards/margins": 7.605309963226318, "rewards/rejected": -12.760688781738281, "step": 770 }, { "epoch": 0.6296670030272452, "grad_norm": 4.545230388641357, "learning_rate": 2.3430237011767166e-06, "logits/chosen": 3.324824810028076, "logits/rejected": 3.0623250007629395, "logps/chosen": -3.5336334705352783, "logps/rejected": -7.56264591217041, "loss": 0.259, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.300450325012207, "rewards/margins": 6.043517589569092, "rewards/rejected": -11.34396743774414, "step": 780 }, { "epoch": 0.6377396569122099, "grad_norm": 3.5774872303009033, "learning_rate": 2.290805391669212e-06, "logits/chosen": 2.70954966545105, "logits/rejected": 2.559739589691162, "logps/chosen": -3.2724106311798096, "logps/rejected": -7.964414119720459, "loss": 0.243, "rewards/accuracies": 0.9375, "rewards/chosen": -4.908616065979004, "rewards/margins": 7.0380048751831055, "rewards/rejected": -11.94662094116211, "step": 790 }, { "epoch": 0.6458123107971746, "grad_norm": 3.9293906688690186, "learning_rate": 2.238678841830867e-06, "logits/chosen": 2.2065200805664062, "logits/rejected": 2.5098512172698975, "logps/chosen": -3.324014186859131, "logps/rejected": -8.072491645812988, "loss": 0.2752, "rewards/accuracies": 0.9125000238418579, "rewards/chosen": -4.986021518707275, "rewards/margins": 7.122715950012207, "rewards/rejected": -12.108736038208008, "step": 800 }, { "epoch": 0.6458123107971746, "eval_logits/chosen": 2.7405736446380615, "eval_logits/rejected": 2.500715732574463, "eval_logps/chosen": -3.4503068923950195, "eval_logps/rejected": -7.3543477058410645, "eval_loss": 0.22608846426010132, "eval_rewards/accuracies": 0.9801980257034302, "eval_rewards/chosen": -5.175460338592529, "eval_rewards/margins": 5.856060981750488, "eval_rewards/rejected": -11.031521797180176, "eval_runtime": 29.0926, "eval_samples_per_second": 27.533, "eval_steps_per_second": 3.472, "step": 800 }, { "epoch": 0.6538849646821393, "grad_norm": 3.1484081745147705, "learning_rate": 2.186666916089239e-06, "logits/chosen": 2.620433807373047, "logits/rejected": 2.554230213165283, "logps/chosen": -3.826986312866211, "logps/rejected": -8.595190048217773, "loss": 0.2267, "rewards/accuracies": 0.9375, "rewards/chosen": -5.740479469299316, "rewards/margins": 7.152307033538818, "rewards/rejected": -12.892786026000977, "step": 810 }, { "epoch": 0.6619576185671039, "grad_norm": 3.712259531021118, "learning_rate": 2.134792428593971e-06, "logits/chosen": 3.1829707622528076, "logits/rejected": 3.0434184074401855, "logps/chosen": -3.3385536670684814, "logps/rejected": -7.571812629699707, "loss": 0.2079, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.0078301429748535, "rewards/margins": 6.349888801574707, "rewards/rejected": -11.357718467712402, "step": 820 }, { "epoch": 0.6700302724520686, "grad_norm": 5.120253086090088, "learning_rate": 2.0830781332097446e-06, "logits/chosen": 3.631582736968994, "logits/rejected": 3.769865036010742, "logps/chosen": -3.2808995246887207, "logps/rejected": -7.530601501464844, "loss": 0.2213, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -4.92134952545166, "rewards/margins": 6.3745527267456055, "rewards/rejected": -11.29590129852295, "step": 830 }, { "epoch": 0.6781029263370333, "grad_norm": 3.723133087158203, "learning_rate": 2.031546713535688e-06, "logits/chosen": 2.7661056518554688, "logits/rejected": 2.7924928665161133, "logps/chosen": -3.5595524311065674, "logps/rejected": -7.967140197753906, "loss": 0.2381, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.339327812194824, "rewards/margins": 6.611382961273193, "rewards/rejected": -11.950711250305176, "step": 840 }, { "epoch": 0.686175580221998, "grad_norm": 3.4145116806030273, "learning_rate": 1.9802207729556023e-06, "logits/chosen": 2.4502460956573486, "logits/rejected": 2.364321708679199, "logps/chosen": -3.2101433277130127, "logps/rejected": -8.211054801940918, "loss": 0.2112, "rewards/accuracies": 0.9624999761581421, "rewards/chosen": -4.81521463394165, "rewards/margins": 7.501368522644043, "rewards/rejected": -12.316584587097168, "step": 850 }, { "epoch": 0.686175580221998, "eval_logits/chosen": 2.7086803913116455, "eval_logits/rejected": 2.437659978866577, "eval_logps/chosen": -3.123488664627075, "eval_logps/rejected": -7.074939727783203, "eval_loss": 0.22627003490924835, "eval_rewards/accuracies": 0.9702970385551453, "eval_rewards/chosen": -4.685232639312744, "eval_rewards/margins": 5.927176475524902, "eval_rewards/rejected": -10.612408638000488, "eval_runtime": 29.0746, "eval_samples_per_second": 27.55, "eval_steps_per_second": 3.474, "step": 850 }, { "epoch": 0.6942482341069627, "grad_norm": 3.628746509552002, "learning_rate": 1.9291228247233607e-06, "logits/chosen": 2.9090940952301025, "logits/rejected": 2.9559710025787354, "logps/chosen": -3.447845458984375, "logps/rejected": -7.511412620544434, "loss": 0.2154, "rewards/accuracies": 0.9375, "rewards/chosen": -5.171767711639404, "rewards/margins": 6.09535026550293, "rewards/rejected": -11.267118453979492, "step": 860 }, { "epoch": 0.7023208879919274, "grad_norm": 3.231868267059326, "learning_rate": 1.8782752820878636e-06, "logits/chosen": 2.6312456130981445, "logits/rejected": 2.458289861679077, "logps/chosen": -3.6256003379821777, "logps/rejected": -8.079771041870117, "loss": 0.2386, "rewards/accuracies": 0.887499988079071, "rewards/chosen": -5.438401699066162, "rewards/margins": 6.681253910064697, "rewards/rejected": -12.11965560913086, "step": 870 }, { "epoch": 0.7103935418768921, "grad_norm": 5.360827445983887, "learning_rate": 1.827700448461836e-06, "logits/chosen": 2.383152961730957, "logits/rejected": 2.461998701095581, "logps/chosen": -3.555072069168091, "logps/rejected": -8.438495635986328, "loss": 0.2126, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.332608222961426, "rewards/margins": 7.32513427734375, "rewards/rejected": -12.657742500305176, "step": 880 }, { "epoch": 0.7184661957618567, "grad_norm": 5.990912914276123, "learning_rate": 1.7774205076388207e-06, "logits/chosen": 3.3776676654815674, "logits/rejected": 2.913318157196045, "logps/chosen": -3.9117507934570312, "logps/rejected": -8.53381633758545, "loss": 0.185, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -5.8676252365112305, "rewards/margins": 6.933099269866943, "rewards/rejected": -12.800724983215332, "step": 890 }, { "epoch": 0.7265388496468214, "grad_norm": 6.092798233032227, "learning_rate": 1.7274575140626318e-06, "logits/chosen": 1.917933702468872, "logits/rejected": 1.6799787282943726, "logps/chosen": -3.4144363403320312, "logps/rejected": -8.597683906555176, "loss": 0.2383, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -5.121654987335205, "rewards/margins": 7.7748703956604, "rewards/rejected": -12.896525382995605, "step": 900 }, { "epoch": 0.7265388496468214, "eval_logits/chosen": 2.6657638549804688, "eval_logits/rejected": 2.4586844444274902, "eval_logps/chosen": -3.4474172592163086, "eval_logps/rejected": -7.506450176239014, "eval_loss": 0.21870465576648712, "eval_rewards/accuracies": 0.9801980257034302, "eval_rewards/chosen": -5.171125888824463, "eval_rewards/margins": 6.088548183441162, "eval_rewards/rejected": -11.259674072265625, "eval_runtime": 29.1008, "eval_samples_per_second": 27.525, "eval_steps_per_second": 3.471, "step": 900 }, { "epoch": 0.7346115035317861, "grad_norm": 3.623443841934204, "learning_rate": 1.677833383153542e-06, "logits/chosen": 2.915825605392456, "logits/rejected": 2.7418575286865234, "logps/chosen": -3.517927646636963, "logps/rejected": -8.27934455871582, "loss": 0.2311, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -5.276891708374023, "rewards/margins": 7.142125129699707, "rewards/rejected": -12.419015884399414, "step": 910 }, { "epoch": 0.7426841574167508, "grad_norm": 4.020949363708496, "learning_rate": 1.6285698816954626e-06, "logits/chosen": 2.4125845432281494, "logits/rejected": 2.6929471492767334, "logps/chosen": -3.646517276763916, "logps/rejected": -7.9424262046813965, "loss": 0.2034, "rewards/accuracies": 0.8999999761581421, "rewards/chosen": -5.469775199890137, "rewards/margins": 6.443863868713379, "rewards/rejected": -11.913639068603516, "step": 920 }, { "epoch": 0.7507568113017155, "grad_norm": 4.215940475463867, "learning_rate": 1.5796886182883053e-06, "logits/chosen": 2.525031566619873, "logits/rejected": 2.478022336959839, "logps/chosen": -3.580151319503784, "logps/rejected": -8.356379508972168, "loss": 0.2301, "rewards/accuracies": 0.9375, "rewards/chosen": -5.370226860046387, "rewards/margins": 7.164343357086182, "rewards/rejected": -12.534570693969727, "step": 930 }, { "epoch": 0.7588294651866802, "grad_norm": 3.3429832458496094, "learning_rate": 1.5312110338697427e-06, "logits/chosen": 2.68546462059021, "logits/rejected": 2.4535439014434814, "logps/chosen": -3.716118574142456, "logps/rejected": -8.706206321716309, "loss": 0.1898, "rewards/accuracies": 0.9375, "rewards/chosen": -5.5741777420043945, "rewards/margins": 7.485130310058594, "rewards/rejected": -13.059308052062988, "step": 940 }, { "epoch": 0.7669021190716448, "grad_norm": 3.5254878997802734, "learning_rate": 1.4831583923105e-06, "logits/chosen": 2.924198627471924, "logits/rejected": 2.7865915298461914, "logps/chosen": -3.7330048084259033, "logps/rejected": -8.79862117767334, "loss": 0.1947, "rewards/accuracies": 0.9375, "rewards/chosen": -5.5995073318481445, "rewards/margins": 7.598424434661865, "rewards/rejected": -13.197932243347168, "step": 950 }, { "epoch": 0.7669021190716448, "eval_logits/chosen": 2.5911405086517334, "eval_logits/rejected": 2.4414517879486084, "eval_logps/chosen": -3.396683692932129, "eval_logps/rejected": -7.610343933105469, "eval_loss": 0.2162380963563919, "eval_rewards/accuracies": 0.9900990128517151, "eval_rewards/chosen": -5.09502649307251, "eval_rewards/margins": 6.32049036026001, "eval_rewards/rejected": -11.415514945983887, "eval_runtime": 29.0842, "eval_samples_per_second": 27.541, "eval_steps_per_second": 3.473, "step": 950 }, { "epoch": 0.7749747729566094, "grad_norm": 2.4495151042938232, "learning_rate": 1.4355517710873184e-06, "logits/chosen": 3.756882429122925, "logits/rejected": 3.461829423904419, "logps/chosen": -3.4112212657928467, "logps/rejected": -7.716673374176025, "loss": 0.1954, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -5.1168317794799805, "rewards/margins": 6.458177089691162, "rewards/rejected": -11.575010299682617, "step": 960 }, { "epoch": 0.7830474268415741, "grad_norm": 3.6960482597351074, "learning_rate": 1.388412052037682e-06, "logits/chosen": 2.683379888534546, "logits/rejected": 2.5084214210510254, "logps/chosen": -3.2802653312683105, "logps/rejected": -8.223787307739258, "loss": 0.1962, "rewards/accuracies": 0.9624999761581421, "rewards/chosen": -4.920398712158203, "rewards/margins": 7.415283203125, "rewards/rejected": -12.33568000793457, "step": 970 }, { "epoch": 0.7911200807265388, "grad_norm": 1.6209157705307007, "learning_rate": 1.3417599122003464e-06, "logits/chosen": 2.2775497436523438, "logits/rejected": 2.4255974292755127, "logps/chosen": -3.296114444732666, "logps/rejected": -7.983359336853027, "loss": 0.1651, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -4.944170951843262, "rewards/margins": 7.030867576599121, "rewards/rejected": -11.975038528442383, "step": 980 }, { "epoch": 0.7991927346115035, "grad_norm": 4.007768154144287, "learning_rate": 1.2956158147457116e-06, "logits/chosen": 3.2154605388641357, "logits/rejected": 2.895961046218872, "logps/chosen": -3.3481571674346924, "logps/rejected": -8.350494384765625, "loss": 0.1702, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -5.022235870361328, "rewards/margins": 7.503505706787109, "rewards/rejected": -12.525741577148438, "step": 990 }, { "epoch": 0.8072653884964682, "grad_norm": 3.026970386505127, "learning_rate": 1.2500000000000007e-06, "logits/chosen": 3.6043121814727783, "logits/rejected": 3.541698932647705, "logps/chosen": -3.722698926925659, "logps/rejected": -8.218320846557617, "loss": 0.1934, "rewards/accuracies": 0.987500011920929, "rewards/chosen": -5.584048271179199, "rewards/margins": 6.743433475494385, "rewards/rejected": -12.327482223510742, "step": 1000 }, { "epoch": 0.8072653884964682, "eval_logits/chosen": 2.707226514816284, "eval_logits/rejected": 2.5214099884033203, "eval_logps/chosen": -3.502831220626831, "eval_logps/rejected": -7.705867767333984, "eval_loss": 0.20982278883457184, "eval_rewards/accuracies": 0.9900990128517151, "eval_rewards/chosen": -5.254247188568115, "eval_rewards/margins": 6.304553985595703, "eval_rewards/rejected": -11.558801651000977, "eval_runtime": 29.0974, "eval_samples_per_second": 27.528, "eval_steps_per_second": 3.471, "step": 1000 }, { "epoch": 0.8153380423814329, "grad_norm": 2.1142029762268066, "learning_rate": 1.204932476567175e-06, "logits/chosen": 3.21277117729187, "logits/rejected": 2.902198314666748, "logps/chosen": -3.9205710887908936, "logps/rejected": -8.282219886779785, "loss": 0.1911, "rewards/accuracies": 0.9375, "rewards/chosen": -5.880856513977051, "rewards/margins": 6.542473793029785, "rewards/rejected": -12.423330307006836, "step": 1010 }, { "epoch": 0.8234106962663976, "grad_norm": 4.392597198486328, "learning_rate": 1.160433012552508e-06, "logits/chosen": 2.785994052886963, "logits/rejected": 2.8172032833099365, "logps/chosen": -3.6294891834259033, "logps/rejected": -8.542880058288574, "loss": 0.2222, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.4442338943481445, "rewards/margins": 7.370086669921875, "rewards/rejected": -12.81432056427002, "step": 1020 }, { "epoch": 0.8314833501513622, "grad_norm": 2.633239269256592, "learning_rate": 1.11652112689164e-06, "logits/chosen": 2.663726329803467, "logits/rejected": 2.4244682788848877, "logps/chosen": -3.625394821166992, "logps/rejected": -8.94005298614502, "loss": 0.1768, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -5.438092231750488, "rewards/margins": 7.971987247467041, "rewards/rejected": -13.410079956054688, "step": 1030 }, { "epoch": 0.8395560040363269, "grad_norm": 2.5595297813415527, "learning_rate": 1.073216080788921e-06, "logits/chosen": 3.432141065597534, "logits/rejected": 3.068664789199829, "logps/chosen": -3.4811911582946777, "logps/rejected": -8.416011810302734, "loss": 0.1923, "rewards/accuracies": 0.949999988079071, "rewards/chosen": -5.221786975860596, "rewards/margins": 7.402231693267822, "rewards/rejected": -12.624017715454102, "step": 1040 }, { "epoch": 0.8476286579212916, "grad_norm": 5.044093132019043, "learning_rate": 1.0305368692688175e-06, "logits/chosen": 2.554561138153076, "logits/rejected": 2.273308753967285, "logps/chosen": -3.726578950881958, "logps/rejected": -8.446795463562012, "loss": 0.1525, "rewards/accuracies": 0.987500011920929, "rewards/chosen": -5.589868545532227, "rewards/margins": 7.080326080322266, "rewards/rejected": -12.670194625854492, "step": 1050 }, { "epoch": 0.8476286579212916, "eval_logits/chosen": 2.738980770111084, "eval_logits/rejected": 2.524784564971924, "eval_logps/chosen": -3.4412803649902344, "eval_logps/rejected": -7.715125560760498, "eval_loss": 0.20868732035160065, "eval_rewards/accuracies": 0.9801980257034302, "eval_rewards/chosen": -5.161920070648193, "eval_rewards/margins": 6.410768032073975, "eval_rewards/rejected": -11.572687149047852, "eval_runtime": 29.0996, "eval_samples_per_second": 27.526, "eval_steps_per_second": 3.471, "step": 1050 }, { "epoch": 0.8557013118062563, "grad_norm": 4.388390064239502, "learning_rate": 9.88502212844063e-07, "logits/chosen": 3.056361198425293, "logits/rejected": 2.8604633808135986, "logps/chosen": -3.4161269664764404, "logps/rejected": -8.637321472167969, "loss": 0.1962, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -5.124190807342529, "rewards/margins": 7.831790924072266, "rewards/rejected": -12.955981254577637, "step": 1060 }, { "epoch": 0.863773965691221, "grad_norm": 5.200846195220947, "learning_rate": 9.471305493042243e-07, "logits/chosen": 2.426769971847534, "logits/rejected": 2.714139461517334, "logps/chosen": -3.3948585987091064, "logps/rejected": -8.400737762451172, "loss": 0.1953, "rewards/accuracies": 0.9750000238418579, "rewards/chosen": -5.092287540435791, "rewards/margins": 7.508819580078125, "rewards/rejected": -12.601107597351074, "step": 1070 }, { "epoch": 0.8718466195761857, "grad_norm": 4.115231513977051, "learning_rate": 9.064400256282757e-07, "logits/chosen": 2.680354356765747, "logits/rejected": 2.4186928272247314, "logps/chosen": -3.406576633453369, "logps/rejected": -8.102071762084961, "loss": 0.1961, "rewards/accuracies": 0.925000011920929, "rewards/chosen": -5.109864711761475, "rewards/margins": 7.04324197769165, "rewards/rejected": -12.153106689453125, "step": 1080 }, { "epoch": 0.8799192734611504, "grad_norm": 2.0617411136627197, "learning_rate": 8.664484900247363e-07, "logits/chosen": 2.8944904804229736, "logits/rejected": 2.763841152191162, "logps/chosen": -3.420555591583252, "logps/rejected": -8.535276412963867, "loss": 0.2107, "rewards/accuracies": 0.9624999761581421, "rewards/chosen": -5.130833625793457, "rewards/margins": 7.672080993652344, "rewards/rejected": -12.802912712097168, "step": 1090 }, { "epoch": 0.887991927346115, "grad_norm": 4.412430763244629, "learning_rate": 8.271734841028553e-07, "logits/chosen": 2.8057053089141846, "logits/rejected": 2.880598545074463, "logps/chosen": -3.727388381958008, "logps/rejected": -8.665179252624512, "loss": 0.2103, "rewards/accuracies": 0.9125000238418579, "rewards/chosen": -5.59108304977417, "rewards/margins": 7.406687259674072, "rewards/rejected": -12.997769355773926, "step": 1100 }, { "epoch": 0.887991927346115, "eval_logits/chosen": 2.784738063812256, "eval_logits/rejected": 2.529378890991211, "eval_logps/chosen": -3.4977424144744873, "eval_logps/rejected": -7.756742477416992, "eval_loss": 0.2044122815132141, "eval_rewards/accuracies": 0.9801980257034302, "eval_rewards/chosen": -5.2466139793396, "eval_rewards/margins": 6.388499736785889, "eval_rewards/rejected": -11.635113716125488, "eval_runtime": 29.0833, "eval_samples_per_second": 27.542, "eval_steps_per_second": 3.473, "step": 1100 } ], "logging_steps": 10, "max_steps": 1500, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 50, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2.5594781195808276e+18, "train_batch_size": 1, "trial_name": null, "trial_params": null }