{ "best_metric": 3.1832265853881836, "best_model_checkpoint": "./models/lora-finetuning/LLaMmlein_120M/checkpoint-118000", "epoch": 2.3397676944931898, "eval_steps": 1000, "global_step": 140000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.008356313194618534, "grad_norm": 40.111167907714844, "learning_rate": 4.96e-05, "loss": 4.8904, "step": 500 }, { "epoch": 0.016712626389237067, "grad_norm": 22.074819564819336, "learning_rate": 4.986173570570655e-05, "loss": 4.4642, "step": 1000 }, { "epoch": 0.016712626389237067, "eval_loss": 4.3232102394104, "eval_runtime": 27.5175, "eval_samples_per_second": 162.152, "eval_steps_per_second": 20.278, "step": 1000 }, { "epoch": 0.025068939583855605, "grad_norm": 19.863744735717773, "learning_rate": 4.972207480237983e-05, "loss": 4.3112, "step": 1500 }, { "epoch": 0.033425252778474135, "grad_norm": 19.56609344482422, "learning_rate": 4.958241389905311e-05, "loss": 4.2377, "step": 2000 }, { "epoch": 0.033425252778474135, "eval_loss": 4.127670764923096, "eval_runtime": 27.7514, "eval_samples_per_second": 160.785, "eval_steps_per_second": 20.107, "step": 2000 }, { "epoch": 0.04178156597309267, "grad_norm": 19.212339401245117, "learning_rate": 4.944275299572638e-05, "loss": 4.1323, "step": 2500 }, { "epoch": 0.05013787916771121, "grad_norm": 16.833269119262695, "learning_rate": 4.930309209239966e-05, "loss": 4.0461, "step": 3000 }, { "epoch": 0.05013787916771121, "eval_loss": 4.040640354156494, "eval_runtime": 27.5887, "eval_samples_per_second": 161.733, "eval_steps_per_second": 20.226, "step": 3000 }, { "epoch": 0.05849419236232974, "grad_norm": 12.117938995361328, "learning_rate": 4.9163431189072935e-05, "loss": 4.0848, "step": 3500 }, { "epoch": 0.06685050555694827, "grad_norm": 16.871612548828125, "learning_rate": 4.9023770285746213e-05, "loss": 4.0421, "step": 4000 }, { "epoch": 0.06685050555694827, "eval_loss": 3.966156005859375, "eval_runtime": 27.5488, "eval_samples_per_second": 161.967, "eval_steps_per_second": 20.255, "step": 4000 }, { "epoch": 0.0752068187515668, "grad_norm": 23.424396514892578, "learning_rate": 4.8884109382419485e-05, "loss": 3.9967, "step": 4500 }, { "epoch": 0.08356313194618534, "grad_norm": 17.837812423706055, "learning_rate": 4.8744448479092763e-05, "loss": 3.9343, "step": 5000 }, { "epoch": 0.08356313194618534, "eval_loss": 3.9122862815856934, "eval_runtime": 27.6483, "eval_samples_per_second": 161.384, "eval_steps_per_second": 20.182, "step": 5000 }, { "epoch": 0.09191944514080387, "grad_norm": 19.62445640563965, "learning_rate": 4.860478757576604e-05, "loss": 3.9297, "step": 5500 }, { "epoch": 0.10027575833542242, "grad_norm": 16.499317169189453, "learning_rate": 4.846512667243932e-05, "loss": 3.8862, "step": 6000 }, { "epoch": 0.10027575833542242, "eval_loss": 3.8697621822357178, "eval_runtime": 30.2568, "eval_samples_per_second": 147.471, "eval_steps_per_second": 18.442, "step": 6000 }, { "epoch": 0.10863207153004095, "grad_norm": 22.481382369995117, "learning_rate": 4.83254657691126e-05, "loss": 3.8479, "step": 6500 }, { "epoch": 0.11698838472465949, "grad_norm": 15.390802383422852, "learning_rate": 4.818580486578587e-05, "loss": 3.8934, "step": 7000 }, { "epoch": 0.11698838472465949, "eval_loss": 3.8397133350372314, "eval_runtime": 27.593, "eval_samples_per_second": 161.708, "eval_steps_per_second": 20.223, "step": 7000 }, { "epoch": 0.12534469791927802, "grad_norm": 16.615388870239258, "learning_rate": 4.804670260607246e-05, "loss": 3.8718, "step": 7500 }, { "epoch": 0.13370101111389654, "grad_norm": 16.377056121826172, "learning_rate": 4.7907041702745736e-05, "loss": 3.813, "step": 8000 }, { "epoch": 0.13370101111389654, "eval_loss": 3.8209779262542725, "eval_runtime": 27.534, "eval_samples_per_second": 162.054, "eval_steps_per_second": 20.266, "step": 8000 }, { "epoch": 0.1420573243085151, "grad_norm": 20.620891571044922, "learning_rate": 4.7767380799419014e-05, "loss": 3.8194, "step": 8500 }, { "epoch": 0.1504136375031336, "grad_norm": 14.788801193237305, "learning_rate": 4.7627719896092286e-05, "loss": 3.8246, "step": 9000 }, { "epoch": 0.1504136375031336, "eval_loss": 3.7902626991271973, "eval_runtime": 27.5751, "eval_samples_per_second": 161.812, "eval_steps_per_second": 20.236, "step": 9000 }, { "epoch": 0.15876995069775215, "grad_norm": 12.923628807067871, "learning_rate": 4.748833831457222e-05, "loss": 3.7739, "step": 9500 }, { "epoch": 0.16712626389237067, "grad_norm": 15.342278480529785, "learning_rate": 4.7348677411245494e-05, "loss": 3.771, "step": 10000 }, { "epoch": 0.16712626389237067, "eval_loss": 3.7538962364196777, "eval_runtime": 27.5894, "eval_samples_per_second": 161.729, "eval_steps_per_second": 20.225, "step": 10000 }, { "epoch": 0.17548257708698922, "grad_norm": 14.153196334838867, "learning_rate": 4.720901650791877e-05, "loss": 3.7321, "step": 10500 }, { "epoch": 0.18383889028160774, "grad_norm": 29.715482711791992, "learning_rate": 4.706935560459205e-05, "loss": 3.7219, "step": 11000 }, { "epoch": 0.18383889028160774, "eval_loss": 3.736612558364868, "eval_runtime": 27.6054, "eval_samples_per_second": 161.635, "eval_steps_per_second": 20.213, "step": 11000 }, { "epoch": 0.1921952034762263, "grad_norm": 22.01296043395996, "learning_rate": 4.692969470126533e-05, "loss": 3.7397, "step": 11500 }, { "epoch": 0.20055151667084484, "grad_norm": 12.3303804397583, "learning_rate": 4.679003379793861e-05, "loss": 3.731, "step": 12000 }, { "epoch": 0.20055151667084484, "eval_loss": 3.7091643810272217, "eval_runtime": 27.5167, "eval_samples_per_second": 162.156, "eval_steps_per_second": 20.279, "step": 12000 }, { "epoch": 0.20890782986546336, "grad_norm": 17.74376106262207, "learning_rate": 4.6650372894611886e-05, "loss": 3.7524, "step": 12500 }, { "epoch": 0.2172641430600819, "grad_norm": 11.327990531921387, "learning_rate": 4.6510711991285164e-05, "loss": 3.6842, "step": 13000 }, { "epoch": 0.2172641430600819, "eval_loss": 3.7014827728271484, "eval_runtime": 27.5638, "eval_samples_per_second": 161.879, "eval_steps_per_second": 20.244, "step": 13000 }, { "epoch": 0.22562045625470042, "grad_norm": 10.342738151550293, "learning_rate": 4.637133040976509e-05, "loss": 3.672, "step": 13500 }, { "epoch": 0.23397676944931897, "grad_norm": 14.019166946411133, "learning_rate": 4.623166950643837e-05, "loss": 3.6743, "step": 14000 }, { "epoch": 0.23397676944931897, "eval_loss": 3.680652618408203, "eval_runtime": 27.7236, "eval_samples_per_second": 160.946, "eval_steps_per_second": 20.127, "step": 14000 }, { "epoch": 0.2423330826439375, "grad_norm": 12.447026252746582, "learning_rate": 4.609200860311165e-05, "loss": 3.68, "step": 14500 }, { "epoch": 0.25068939583855604, "grad_norm": 16.875900268554688, "learning_rate": 4.595234769978493e-05, "loss": 3.6755, "step": 15000 }, { "epoch": 0.25068939583855604, "eval_loss": 3.656320095062256, "eval_runtime": 27.5417, "eval_samples_per_second": 162.009, "eval_steps_per_second": 20.26, "step": 15000 }, { "epoch": 0.2590457090331746, "grad_norm": 14.400089263916016, "learning_rate": 4.581268679645821e-05, "loss": 3.668, "step": 15500 }, { "epoch": 0.2674020222277931, "grad_norm": 17.918594360351562, "learning_rate": 4.567302589313148e-05, "loss": 3.6105, "step": 16000 }, { "epoch": 0.2674020222277931, "eval_loss": 3.6530709266662598, "eval_runtime": 27.5371, "eval_samples_per_second": 162.036, "eval_steps_per_second": 20.264, "step": 16000 }, { "epoch": 0.2757583354224116, "grad_norm": 25.155494689941406, "learning_rate": 4.5533644311611415e-05, "loss": 3.6537, "step": 16500 }, { "epoch": 0.2841146486170302, "grad_norm": 19.406333923339844, "learning_rate": 4.5393983408284686e-05, "loss": 3.6321, "step": 17000 }, { "epoch": 0.2841146486170302, "eval_loss": 3.625420331954956, "eval_runtime": 27.6545, "eval_samples_per_second": 161.348, "eval_steps_per_second": 20.178, "step": 17000 }, { "epoch": 0.2924709618116487, "grad_norm": 16.296859741210938, "learning_rate": 4.5254322504957965e-05, "loss": 3.5798, "step": 17500 }, { "epoch": 0.3008272750062672, "grad_norm": 14.436009407043457, "learning_rate": 4.511466160163124e-05, "loss": 3.5772, "step": 18000 }, { "epoch": 0.3008272750062672, "eval_loss": 3.609403371810913, "eval_runtime": 27.6116, "eval_samples_per_second": 161.599, "eval_steps_per_second": 20.209, "step": 18000 }, { "epoch": 0.30918358820088576, "grad_norm": 14.975789070129395, "learning_rate": 4.497500069830452e-05, "loss": 3.5887, "step": 18500 }, { "epoch": 0.3175399013955043, "grad_norm": 13.919900894165039, "learning_rate": 4.48353397949778e-05, "loss": 3.5855, "step": 19000 }, { "epoch": 0.3175399013955043, "eval_loss": 3.593752384185791, "eval_runtime": 27.5247, "eval_samples_per_second": 162.109, "eval_steps_per_second": 20.273, "step": 19000 }, { "epoch": 0.32589621459012286, "grad_norm": 13.113483428955078, "learning_rate": 4.469567889165107e-05, "loss": 3.5958, "step": 19500 }, { "epoch": 0.33425252778474135, "grad_norm": 16.182727813720703, "learning_rate": 4.455601798832435e-05, "loss": 3.5544, "step": 20000 }, { "epoch": 0.33425252778474135, "eval_loss": 3.5966029167175293, "eval_runtime": 27.5874, "eval_samples_per_second": 161.741, "eval_steps_per_second": 20.227, "step": 20000 }, { "epoch": 0.3426088409793599, "grad_norm": 11.520635604858398, "learning_rate": 4.441635708499763e-05, "loss": 3.5985, "step": 20500 }, { "epoch": 0.35096515417397844, "grad_norm": 9.132452011108398, "learning_rate": 4.4276696181670906e-05, "loss": 3.5385, "step": 21000 }, { "epoch": 0.35096515417397844, "eval_loss": 3.5922670364379883, "eval_runtime": 27.5592, "eval_samples_per_second": 161.906, "eval_steps_per_second": 20.247, "step": 21000 }, { "epoch": 0.359321467368597, "grad_norm": 11.077112197875977, "learning_rate": 4.413703527834418e-05, "loss": 3.5502, "step": 21500 }, { "epoch": 0.3676777805632155, "grad_norm": 10.871437072753906, "learning_rate": 4.3997653696824114e-05, "loss": 3.5683, "step": 22000 }, { "epoch": 0.3676777805632155, "eval_loss": 3.570159673690796, "eval_runtime": 27.5952, "eval_samples_per_second": 161.695, "eval_steps_per_second": 20.221, "step": 22000 }, { "epoch": 0.37603409375783403, "grad_norm": 14.359475135803223, "learning_rate": 4.3858272115304044e-05, "loss": 3.6252, "step": 22500 }, { "epoch": 0.3843904069524526, "grad_norm": 14.215445518493652, "learning_rate": 4.371861121197732e-05, "loss": 3.5357, "step": 23000 }, { "epoch": 0.3843904069524526, "eval_loss": 3.5666186809539795, "eval_runtime": 27.4876, "eval_samples_per_second": 162.328, "eval_steps_per_second": 20.3, "step": 23000 }, { "epoch": 0.3927467201470711, "grad_norm": 12.356164932250977, "learning_rate": 4.3578950308650594e-05, "loss": 3.5561, "step": 23500 }, { "epoch": 0.4011030333416897, "grad_norm": 13.857044219970703, "learning_rate": 4.343956872713053e-05, "loss": 3.5108, "step": 24000 }, { "epoch": 0.4011030333416897, "eval_loss": 3.553119659423828, "eval_runtime": 27.5959, "eval_samples_per_second": 161.691, "eval_steps_per_second": 20.22, "step": 24000 }, { "epoch": 0.40945934653630817, "grad_norm": 18.891963958740234, "learning_rate": 4.329990782380381e-05, "loss": 3.5144, "step": 24500 }, { "epoch": 0.4178156597309267, "grad_norm": 11.144295692443848, "learning_rate": 4.316024692047708e-05, "loss": 3.4772, "step": 25000 }, { "epoch": 0.4178156597309267, "eval_loss": 3.539018154144287, "eval_runtime": 27.6102, "eval_samples_per_second": 161.607, "eval_steps_per_second": 20.21, "step": 25000 }, { "epoch": 0.42617197292554526, "grad_norm": 13.266519546508789, "learning_rate": 4.302058601715036e-05, "loss": 3.4983, "step": 25500 }, { "epoch": 0.4345282861201638, "grad_norm": 13.068937301635742, "learning_rate": 4.288092511382364e-05, "loss": 3.5372, "step": 26000 }, { "epoch": 0.4345282861201638, "eval_loss": 3.5329203605651855, "eval_runtime": 27.5669, "eval_samples_per_second": 161.861, "eval_steps_per_second": 20.242, "step": 26000 }, { "epoch": 0.4428845993147823, "grad_norm": 10.656412124633789, "learning_rate": 4.2741543532303566e-05, "loss": 3.4773, "step": 26500 }, { "epoch": 0.45124091250940085, "grad_norm": 13.304460525512695, "learning_rate": 4.2601882628976845e-05, "loss": 3.5119, "step": 27000 }, { "epoch": 0.45124091250940085, "eval_loss": 3.5271201133728027, "eval_runtime": 27.5897, "eval_samples_per_second": 161.727, "eval_steps_per_second": 20.225, "step": 27000 }, { "epoch": 0.4595972257040194, "grad_norm": 9.64598560333252, "learning_rate": 4.246222172565012e-05, "loss": 3.4904, "step": 27500 }, { "epoch": 0.46795353889863794, "grad_norm": 18.846969604492188, "learning_rate": 4.23225608223234e-05, "loss": 3.4627, "step": 28000 }, { "epoch": 0.46795353889863794, "eval_loss": 3.52040433883667, "eval_runtime": 27.5839, "eval_samples_per_second": 161.761, "eval_steps_per_second": 20.229, "step": 28000 }, { "epoch": 0.47630985209325644, "grad_norm": 11.906749725341797, "learning_rate": 4.218289991899668e-05, "loss": 3.4874, "step": 28500 }, { "epoch": 0.484666165287875, "grad_norm": 11.39976692199707, "learning_rate": 4.204323901566996e-05, "loss": 3.5089, "step": 29000 }, { "epoch": 0.484666165287875, "eval_loss": 3.5191400051116943, "eval_runtime": 27.5537, "eval_samples_per_second": 161.938, "eval_steps_per_second": 20.251, "step": 29000 }, { "epoch": 0.49302247848249353, "grad_norm": 16.590219497680664, "learning_rate": 4.190385743414989e-05, "loss": 3.4669, "step": 29500 }, { "epoch": 0.5013787916771121, "grad_norm": 12.839526176452637, "learning_rate": 4.1764196530823166e-05, "loss": 3.5124, "step": 30000 }, { "epoch": 0.5013787916771121, "eval_loss": 3.5098681449890137, "eval_runtime": 27.52, "eval_samples_per_second": 162.137, "eval_steps_per_second": 20.276, "step": 30000 }, { "epoch": 0.5097351048717306, "grad_norm": 15.809226989746094, "learning_rate": 4.1624535627496444e-05, "loss": 3.4721, "step": 30500 }, { "epoch": 0.5180914180663492, "grad_norm": 14.224934577941895, "learning_rate": 4.148487472416972e-05, "loss": 3.4529, "step": 31000 }, { "epoch": 0.5180914180663492, "eval_loss": 3.4936439990997314, "eval_runtime": 27.5971, "eval_samples_per_second": 161.684, "eval_steps_per_second": 20.22, "step": 31000 }, { "epoch": 0.5264477312609677, "grad_norm": 14.624770164489746, "learning_rate": 4.1345213820842994e-05, "loss": 3.4888, "step": 31500 }, { "epoch": 0.5348040444555862, "grad_norm": 11.033506393432617, "learning_rate": 4.120555291751627e-05, "loss": 3.4595, "step": 32000 }, { "epoch": 0.5348040444555862, "eval_loss": 3.4841041564941406, "eval_runtime": 27.5814, "eval_samples_per_second": 161.776, "eval_steps_per_second": 20.231, "step": 32000 }, { "epoch": 0.5431603576502048, "grad_norm": 11.615704536437988, "learning_rate": 4.106589201418955e-05, "loss": 3.4677, "step": 32500 }, { "epoch": 0.5515166708448233, "grad_norm": 13.639912605285645, "learning_rate": 4.092651043266948e-05, "loss": 3.4418, "step": 33000 }, { "epoch": 0.5515166708448233, "eval_loss": 3.478278398513794, "eval_runtime": 27.6018, "eval_samples_per_second": 161.656, "eval_steps_per_second": 20.216, "step": 33000 }, { "epoch": 0.5598729840394417, "grad_norm": 17.30266761779785, "learning_rate": 4.078684952934276e-05, "loss": 3.4707, "step": 33500 }, { "epoch": 0.5682292972340603, "grad_norm": 10.405800819396973, "learning_rate": 4.064718862601604e-05, "loss": 3.4683, "step": 34000 }, { "epoch": 0.5682292972340603, "eval_loss": 3.474083662033081, "eval_runtime": 27.598, "eval_samples_per_second": 161.678, "eval_steps_per_second": 20.219, "step": 34000 }, { "epoch": 0.5765856104286788, "grad_norm": 12.891817092895508, "learning_rate": 4.0507527722689315e-05, "loss": 3.4564, "step": 34500 }, { "epoch": 0.5849419236232974, "grad_norm": 8.831328392028809, "learning_rate": 4.036786681936259e-05, "loss": 3.3926, "step": 35000 }, { "epoch": 0.5849419236232974, "eval_loss": 3.4699606895446777, "eval_runtime": 27.6266, "eval_samples_per_second": 161.511, "eval_steps_per_second": 20.198, "step": 35000 }, { "epoch": 0.5932982368179159, "grad_norm": 16.12383460998535, "learning_rate": 4.0228205916035865e-05, "loss": 3.4639, "step": 35500 }, { "epoch": 0.6016545500125344, "grad_norm": 12.538627624511719, "learning_rate": 4.0088545012709144e-05, "loss": 3.4289, "step": 36000 }, { "epoch": 0.6016545500125344, "eval_loss": 3.4501824378967285, "eval_runtime": 27.6054, "eval_samples_per_second": 161.635, "eval_steps_per_second": 20.213, "step": 36000 }, { "epoch": 0.610010863207153, "grad_norm": 13.25362777709961, "learning_rate": 3.994888410938242e-05, "loss": 3.4552, "step": 36500 }, { "epoch": 0.6183671764017715, "grad_norm": 14.144664764404297, "learning_rate": 3.980950252786235e-05, "loss": 3.432, "step": 37000 }, { "epoch": 0.6183671764017715, "eval_loss": 3.4386274814605713, "eval_runtime": 27.5191, "eval_samples_per_second": 162.142, "eval_steps_per_second": 20.277, "step": 37000 }, { "epoch": 0.6267234895963901, "grad_norm": 11.082966804504395, "learning_rate": 3.966984162453563e-05, "loss": 3.446, "step": 37500 }, { "epoch": 0.6350798027910086, "grad_norm": 12.105545997619629, "learning_rate": 3.953018072120891e-05, "loss": 3.4498, "step": 38000 }, { "epoch": 0.6350798027910086, "eval_loss": 3.4284586906433105, "eval_runtime": 27.6228, "eval_samples_per_second": 161.533, "eval_steps_per_second": 20.201, "step": 38000 }, { "epoch": 0.6434361159856271, "grad_norm": 11.420595169067383, "learning_rate": 3.939051981788218e-05, "loss": 3.3634, "step": 38500 }, { "epoch": 0.6517924291802457, "grad_norm": 13.421010971069336, "learning_rate": 3.9251138236362116e-05, "loss": 3.4011, "step": 39000 }, { "epoch": 0.6517924291802457, "eval_loss": 3.4333345890045166, "eval_runtime": 27.6141, "eval_samples_per_second": 161.584, "eval_steps_per_second": 20.207, "step": 39000 }, { "epoch": 0.6601487423748642, "grad_norm": 13.694308280944824, "learning_rate": 3.911147733303539e-05, "loss": 3.3544, "step": 39500 }, { "epoch": 0.6685050555694827, "grad_norm": 13.354585647583008, "learning_rate": 3.8971816429708666e-05, "loss": 3.4279, "step": 40000 }, { "epoch": 0.6685050555694827, "eval_loss": 3.428675651550293, "eval_runtime": 29.3177, "eval_samples_per_second": 152.194, "eval_steps_per_second": 19.033, "step": 40000 }, { "epoch": 0.6768613687641013, "grad_norm": 13.542250633239746, "learning_rate": 3.8832155526381945e-05, "loss": 3.3683, "step": 40500 }, { "epoch": 0.6852176819587198, "grad_norm": 15.259937286376953, "learning_rate": 3.869249462305522e-05, "loss": 3.3964, "step": 41000 }, { "epoch": 0.6852176819587198, "eval_loss": 3.4252407550811768, "eval_runtime": 32.0858, "eval_samples_per_second": 139.065, "eval_steps_per_second": 17.391, "step": 41000 }, { "epoch": 0.6935739951533384, "grad_norm": 8.693069458007812, "learning_rate": 3.85528337197285e-05, "loss": 3.3419, "step": 41500 }, { "epoch": 0.7019303083479569, "grad_norm": 10.193922996520996, "learning_rate": 3.841317281640178e-05, "loss": 3.361, "step": 42000 }, { "epoch": 0.7019303083479569, "eval_loss": 3.423677444458008, "eval_runtime": 27.5904, "eval_samples_per_second": 161.723, "eval_steps_per_second": 20.224, "step": 42000 }, { "epoch": 0.7102866215425754, "grad_norm": 13.626117706298828, "learning_rate": 3.827351191307506e-05, "loss": 3.3456, "step": 42500 }, { "epoch": 0.718642934737194, "grad_norm": 15.671127319335938, "learning_rate": 3.813413033155499e-05, "loss": 3.39, "step": 43000 }, { "epoch": 0.718642934737194, "eval_loss": 3.410151243209839, "eval_runtime": 27.586, "eval_samples_per_second": 161.749, "eval_steps_per_second": 20.228, "step": 43000 }, { "epoch": 0.7269992479318125, "grad_norm": 13.179340362548828, "learning_rate": 3.7994469428228266e-05, "loss": 3.3407, "step": 43500 }, { "epoch": 0.735355561126431, "grad_norm": 11.219006538391113, "learning_rate": 3.7854808524901544e-05, "loss": 3.3509, "step": 44000 }, { "epoch": 0.735355561126431, "eval_loss": 3.4000790119171143, "eval_runtime": 27.5216, "eval_samples_per_second": 162.127, "eval_steps_per_second": 20.275, "step": 44000 }, { "epoch": 0.7437118743210496, "grad_norm": 19.535215377807617, "learning_rate": 3.771514762157482e-05, "loss": 3.3784, "step": 44500 }, { "epoch": 0.7520681875156681, "grad_norm": 11.256051063537598, "learning_rate": 3.7575486718248094e-05, "loss": 3.3584, "step": 45000 }, { "epoch": 0.7520681875156681, "eval_loss": 3.3937857151031494, "eval_runtime": 27.6288, "eval_samples_per_second": 161.498, "eval_steps_per_second": 20.196, "step": 45000 }, { "epoch": 0.7604245007102867, "grad_norm": 14.835156440734863, "learning_rate": 3.743610513672803e-05, "loss": 3.3964, "step": 45500 }, { "epoch": 0.7687808139049052, "grad_norm": 13.36843204498291, "learning_rate": 3.72964442334013e-05, "loss": 3.3612, "step": 46000 }, { "epoch": 0.7687808139049052, "eval_loss": 3.4002346992492676, "eval_runtime": 27.6951, "eval_samples_per_second": 161.112, "eval_steps_per_second": 20.148, "step": 46000 }, { "epoch": 0.7771371270995237, "grad_norm": 15.822681427001953, "learning_rate": 3.715706265188124e-05, "loss": 3.3235, "step": 46500 }, { "epoch": 0.7854934402941423, "grad_norm": 11.626577377319336, "learning_rate": 3.701740174855451e-05, "loss": 3.3167, "step": 47000 }, { "epoch": 0.7854934402941423, "eval_loss": 3.3830487728118896, "eval_runtime": 27.7333, "eval_samples_per_second": 160.89, "eval_steps_per_second": 20.12, "step": 47000 }, { "epoch": 0.7938497534887607, "grad_norm": 18.387489318847656, "learning_rate": 3.687774084522779e-05, "loss": 3.3468, "step": 47500 }, { "epoch": 0.8022060666833793, "grad_norm": 10.468737602233887, "learning_rate": 3.673807994190107e-05, "loss": 3.3765, "step": 48000 }, { "epoch": 0.8022060666833793, "eval_loss": 3.3805253505706787, "eval_runtime": 27.5567, "eval_samples_per_second": 161.921, "eval_steps_per_second": 20.249, "step": 48000 }, { "epoch": 0.8105623798779978, "grad_norm": 12.263431549072266, "learning_rate": 3.6598419038574345e-05, "loss": 3.3353, "step": 48500 }, { "epoch": 0.8189186930726163, "grad_norm": 10.66336441040039, "learning_rate": 3.6458758135247623e-05, "loss": 3.2779, "step": 49000 }, { "epoch": 0.8189186930726163, "eval_loss": 3.3863792419433594, "eval_runtime": 27.5626, "eval_samples_per_second": 161.886, "eval_steps_per_second": 20.245, "step": 49000 }, { "epoch": 0.8272750062672349, "grad_norm": 13.781414031982422, "learning_rate": 3.6319097231920895e-05, "loss": 3.3591, "step": 49500 }, { "epoch": 0.8356313194618534, "grad_norm": 10.030421257019043, "learning_rate": 3.617943632859417e-05, "loss": 3.3354, "step": 50000 }, { "epoch": 0.8356313194618534, "eval_loss": 3.367990493774414, "eval_runtime": 27.5956, "eval_samples_per_second": 161.693, "eval_steps_per_second": 20.221, "step": 50000 }, { "epoch": 0.8439876326564719, "grad_norm": 14.890128135681152, "learning_rate": 3.603977542526745e-05, "loss": 3.3061, "step": 50500 }, { "epoch": 0.8523439458510905, "grad_norm": 16.298643112182617, "learning_rate": 3.590011452194073e-05, "loss": 3.3032, "step": 51000 }, { "epoch": 0.8523439458510905, "eval_loss": 3.3711585998535156, "eval_runtime": 27.5103, "eval_samples_per_second": 162.194, "eval_steps_per_second": 20.283, "step": 51000 }, { "epoch": 0.860700259045709, "grad_norm": 11.454365730285645, "learning_rate": 3.576045361861401e-05, "loss": 3.2919, "step": 51500 }, { "epoch": 0.8690565722403276, "grad_norm": 11.1732177734375, "learning_rate": 3.562107203709394e-05, "loss": 3.2936, "step": 52000 }, { "epoch": 0.8690565722403276, "eval_loss": 3.358672857284546, "eval_runtime": 27.5316, "eval_samples_per_second": 162.068, "eval_steps_per_second": 20.268, "step": 52000 }, { "epoch": 0.8774128854349461, "grad_norm": 12.489287376403809, "learning_rate": 3.5481411133767216e-05, "loss": 3.2694, "step": 52500 }, { "epoch": 0.8857691986295646, "grad_norm": 12.570661544799805, "learning_rate": 3.534175023044049e-05, "loss": 3.322, "step": 53000 }, { "epoch": 0.8857691986295646, "eval_loss": 3.3604071140289307, "eval_runtime": 27.5745, "eval_samples_per_second": 161.816, "eval_steps_per_second": 20.236, "step": 53000 }, { "epoch": 0.8941255118241832, "grad_norm": 17.960376739501953, "learning_rate": 3.5202368648920424e-05, "loss": 3.2955, "step": 53500 }, { "epoch": 0.9024818250188017, "grad_norm": 13.333609580993652, "learning_rate": 3.5062707745593696e-05, "loss": 3.3394, "step": 54000 }, { "epoch": 0.9024818250188017, "eval_loss": 3.3498120307922363, "eval_runtime": 27.561, "eval_samples_per_second": 161.896, "eval_steps_per_second": 20.246, "step": 54000 }, { "epoch": 0.9108381382134202, "grad_norm": 16.366514205932617, "learning_rate": 3.4923046842266974e-05, "loss": 3.3223, "step": 54500 }, { "epoch": 0.9191944514080388, "grad_norm": 10.783904075622559, "learning_rate": 3.478338593894025e-05, "loss": 3.2717, "step": 55000 }, { "epoch": 0.9191944514080388, "eval_loss": 3.3506462574005127, "eval_runtime": 27.5889, "eval_samples_per_second": 161.732, "eval_steps_per_second": 20.226, "step": 55000 }, { "epoch": 0.9275507646026573, "grad_norm": 12.693829536437988, "learning_rate": 3.464372503561353e-05, "loss": 3.2696, "step": 55500 }, { "epoch": 0.9359070777972759, "grad_norm": 20.29674530029297, "learning_rate": 3.450406413228681e-05, "loss": 3.3342, "step": 56000 }, { "epoch": 0.9359070777972759, "eval_loss": 3.333944797515869, "eval_runtime": 27.579, "eval_samples_per_second": 161.79, "eval_steps_per_second": 20.233, "step": 56000 }, { "epoch": 0.9442633909918944, "grad_norm": 14.309937477111816, "learning_rate": 3.436440322896009e-05, "loss": 3.3321, "step": 56500 }, { "epoch": 0.9526197041865129, "grad_norm": 8.43278980255127, "learning_rate": 3.4224742325633366e-05, "loss": 3.2396, "step": 57000 }, { "epoch": 0.9526197041865129, "eval_loss": 3.3314433097839355, "eval_runtime": 27.6061, "eval_samples_per_second": 161.631, "eval_steps_per_second": 20.213, "step": 57000 }, { "epoch": 0.9609760173811315, "grad_norm": 10.31540584564209, "learning_rate": 3.4085081422306644e-05, "loss": 3.2436, "step": 57500 }, { "epoch": 0.96933233057575, "grad_norm": 10.261503219604492, "learning_rate": 3.3945699840786574e-05, "loss": 3.2845, "step": 58000 }, { "epoch": 0.96933233057575, "eval_loss": 3.3237485885620117, "eval_runtime": 27.545, "eval_samples_per_second": 161.99, "eval_steps_per_second": 20.258, "step": 58000 }, { "epoch": 0.9776886437703685, "grad_norm": 10.157827377319336, "learning_rate": 3.380603893745985e-05, "loss": 3.2976, "step": 58500 }, { "epoch": 0.9860449569649871, "grad_norm": 12.794463157653809, "learning_rate": 3.366637803413313e-05, "loss": 3.2621, "step": 59000 }, { "epoch": 0.9860449569649871, "eval_loss": 3.325364351272583, "eval_runtime": 27.6749, "eval_samples_per_second": 161.229, "eval_steps_per_second": 20.163, "step": 59000 }, { "epoch": 0.9944012701596056, "grad_norm": 17.333826065063477, "learning_rate": 3.35267171308064e-05, "loss": 3.2696, "step": 59500 }, { "epoch": 1.0027575833542242, "grad_norm": 12.315380096435547, "learning_rate": 3.338705622747968e-05, "loss": 3.2115, "step": 60000 }, { "epoch": 1.0027575833542242, "eval_loss": 3.336367607116699, "eval_runtime": 27.5799, "eval_samples_per_second": 161.785, "eval_steps_per_second": 20.232, "step": 60000 }, { "epoch": 1.0111138965488426, "grad_norm": 23.34908676147461, "learning_rate": 3.324739532415296e-05, "loss": 2.9589, "step": 60500 }, { "epoch": 1.0194702097434611, "grad_norm": 10.89417839050293, "learning_rate": 3.310773442082624e-05, "loss": 3.0302, "step": 61000 }, { "epoch": 1.0194702097434611, "eval_loss": 3.325634479522705, "eval_runtime": 27.573, "eval_samples_per_second": 161.825, "eval_steps_per_second": 20.237, "step": 61000 }, { "epoch": 1.0278265229380796, "grad_norm": 16.59639549255371, "learning_rate": 3.296835283930617e-05, "loss": 2.9884, "step": 61500 }, { "epoch": 1.0361828361326983, "grad_norm": 13.3978910446167, "learning_rate": 3.2828691935979445e-05, "loss": 2.9762, "step": 62000 }, { "epoch": 1.0361828361326983, "eval_loss": 3.334028482437134, "eval_runtime": 27.5551, "eval_samples_per_second": 161.93, "eval_steps_per_second": 20.25, "step": 62000 }, { "epoch": 1.0445391493273168, "grad_norm": 10.937264442443848, "learning_rate": 3.2689031032652723e-05, "loss": 2.9597, "step": 62500 }, { "epoch": 1.0528954625219353, "grad_norm": 14.150084495544434, "learning_rate": 3.2549370129325995e-05, "loss": 2.997, "step": 63000 }, { "epoch": 1.0528954625219353, "eval_loss": 3.320002794265747, "eval_runtime": 27.5918, "eval_samples_per_second": 161.715, "eval_steps_per_second": 20.223, "step": 63000 }, { "epoch": 1.0612517757165538, "grad_norm": 10.700261116027832, "learning_rate": 3.240970922599927e-05, "loss": 2.9857, "step": 63500 }, { "epoch": 1.0696080889111723, "grad_norm": 11.004881858825684, "learning_rate": 3.22703276444792e-05, "loss": 2.9591, "step": 64000 }, { "epoch": 1.0696080889111723, "eval_loss": 3.333744525909424, "eval_runtime": 27.6282, "eval_samples_per_second": 161.502, "eval_steps_per_second": 20.197, "step": 64000 }, { "epoch": 1.077964402105791, "grad_norm": 10.794275283813477, "learning_rate": 3.213066674115248e-05, "loss": 2.9817, "step": 64500 }, { "epoch": 1.0863207153004095, "grad_norm": 15.817968368530273, "learning_rate": 3.199100583782576e-05, "loss": 2.9543, "step": 65000 }, { "epoch": 1.0863207153004095, "eval_loss": 3.3309056758880615, "eval_runtime": 27.5459, "eval_samples_per_second": 161.984, "eval_steps_per_second": 20.257, "step": 65000 }, { "epoch": 1.094677028495028, "grad_norm": 14.550418853759766, "learning_rate": 3.185134493449904e-05, "loss": 2.9485, "step": 65500 }, { "epoch": 1.1030333416896465, "grad_norm": 11.362966537475586, "learning_rate": 3.1711684031172316e-05, "loss": 2.9787, "step": 66000 }, { "epoch": 1.1030333416896465, "eval_loss": 3.332648992538452, "eval_runtime": 27.619, "eval_samples_per_second": 161.555, "eval_steps_per_second": 20.203, "step": 66000 }, { "epoch": 1.111389654884265, "grad_norm": 14.36471176147461, "learning_rate": 3.157202312784559e-05, "loss": 2.9943, "step": 66500 }, { "epoch": 1.1197459680788837, "grad_norm": 17.348573684692383, "learning_rate": 3.1432641546325524e-05, "loss": 3.033, "step": 67000 }, { "epoch": 1.1197459680788837, "eval_loss": 3.311136245727539, "eval_runtime": 27.6024, "eval_samples_per_second": 161.653, "eval_steps_per_second": 20.216, "step": 67000 }, { "epoch": 1.1281022812735022, "grad_norm": 13.361127853393555, "learning_rate": 3.1292980642998796e-05, "loss": 2.995, "step": 67500 }, { "epoch": 1.1364585944681207, "grad_norm": 12.931785583496094, "learning_rate": 3.1153319739672074e-05, "loss": 2.9679, "step": 68000 }, { "epoch": 1.1364585944681207, "eval_loss": 3.308124542236328, "eval_runtime": 27.5871, "eval_samples_per_second": 161.742, "eval_steps_per_second": 20.227, "step": 68000 }, { "epoch": 1.1448149076627392, "grad_norm": 15.317282676696777, "learning_rate": 3.101365883634535e-05, "loss": 3.0068, "step": 68500 }, { "epoch": 1.1531712208573577, "grad_norm": 16.179967880249023, "learning_rate": 3.087399793301863e-05, "loss": 2.9658, "step": 69000 }, { "epoch": 1.1531712208573577, "eval_loss": 3.3181824684143066, "eval_runtime": 27.6733, "eval_samples_per_second": 161.238, "eval_steps_per_second": 20.164, "step": 69000 }, { "epoch": 1.1615275340519762, "grad_norm": 15.436213493347168, "learning_rate": 3.073433702969191e-05, "loss": 3.0074, "step": 69500 }, { "epoch": 1.1698838472465949, "grad_norm": 27.164413452148438, "learning_rate": 3.059467612636519e-05, "loss": 2.9649, "step": 70000 }, { "epoch": 1.1698838472465949, "eval_loss": 3.3080978393554688, "eval_runtime": 27.6434, "eval_samples_per_second": 161.413, "eval_steps_per_second": 20.186, "step": 70000 }, { "epoch": 1.1782401604412134, "grad_norm": 11.414698600769043, "learning_rate": 3.045529454484512e-05, "loss": 3.0125, "step": 70500 }, { "epoch": 1.1865964736358319, "grad_norm": 15.268623352050781, "learning_rate": 3.0315633641518392e-05, "loss": 2.9853, "step": 71000 }, { "epoch": 1.1865964736358319, "eval_loss": 3.298069477081299, "eval_runtime": 27.61, "eval_samples_per_second": 161.608, "eval_steps_per_second": 20.21, "step": 71000 }, { "epoch": 1.1949527868304504, "grad_norm": 23.319032669067383, "learning_rate": 3.017625205999833e-05, "loss": 2.9738, "step": 71500 }, { "epoch": 1.2033091000250689, "grad_norm": 11.64974308013916, "learning_rate": 3.00365911566716e-05, "loss": 2.9607, "step": 72000 }, { "epoch": 1.2033091000250689, "eval_loss": 3.3039419651031494, "eval_runtime": 29.6202, "eval_samples_per_second": 150.64, "eval_steps_per_second": 18.838, "step": 72000 }, { "epoch": 1.2116654132196876, "grad_norm": 11.017394065856934, "learning_rate": 2.989693025334488e-05, "loss": 2.9694, "step": 72500 }, { "epoch": 1.220021726414306, "grad_norm": 11.3243989944458, "learning_rate": 2.9757269350018157e-05, "loss": 2.9665, "step": 73000 }, { "epoch": 1.220021726414306, "eval_loss": 3.302910804748535, "eval_runtime": 27.6122, "eval_samples_per_second": 161.595, "eval_steps_per_second": 20.208, "step": 73000 }, { "epoch": 1.2283780396089246, "grad_norm": 14.27160358428955, "learning_rate": 2.9617608446691435e-05, "loss": 2.9554, "step": 73500 }, { "epoch": 1.236734352803543, "grad_norm": 9.526435852050781, "learning_rate": 2.9478226865171365e-05, "loss": 3.0167, "step": 74000 }, { "epoch": 1.236734352803543, "eval_loss": 3.3012630939483643, "eval_runtime": 27.614, "eval_samples_per_second": 161.584, "eval_steps_per_second": 20.207, "step": 74000 }, { "epoch": 1.2450906659981615, "grad_norm": 14.875115394592285, "learning_rate": 2.9338565961844643e-05, "loss": 3.0263, "step": 74500 }, { "epoch": 1.25344697919278, "grad_norm": 16.816545486450195, "learning_rate": 2.919890505851792e-05, "loss": 2.9977, "step": 75000 }, { "epoch": 1.25344697919278, "eval_loss": 3.3035476207733154, "eval_runtime": 27.574, "eval_samples_per_second": 161.819, "eval_steps_per_second": 20.236, "step": 75000 }, { "epoch": 1.2618032923873987, "grad_norm": 16.662649154663086, "learning_rate": 2.9059244155191196e-05, "loss": 2.9594, "step": 75500 }, { "epoch": 1.2701596055820172, "grad_norm": 14.543773651123047, "learning_rate": 2.8919583251864475e-05, "loss": 2.9845, "step": 76000 }, { "epoch": 1.2701596055820172, "eval_loss": 3.302872896194458, "eval_runtime": 27.5299, "eval_samples_per_second": 162.078, "eval_steps_per_second": 20.269, "step": 76000 }, { "epoch": 1.2785159187766357, "grad_norm": 15.129777908325195, "learning_rate": 2.8779922348537753e-05, "loss": 2.9826, "step": 76500 }, { "epoch": 1.2868722319712542, "grad_norm": 13.58123779296875, "learning_rate": 2.864026144521103e-05, "loss": 2.9302, "step": 77000 }, { "epoch": 1.2868722319712542, "eval_loss": 3.287860155105591, "eval_runtime": 27.5656, "eval_samples_per_second": 161.868, "eval_steps_per_second": 20.243, "step": 77000 }, { "epoch": 1.2952285451658727, "grad_norm": 13.634276390075684, "learning_rate": 2.8500600541884303e-05, "loss": 2.9802, "step": 77500 }, { "epoch": 1.3035848583604914, "grad_norm": 12.221925735473633, "learning_rate": 2.836093963855758e-05, "loss": 3.0119, "step": 78000 }, { "epoch": 1.3035848583604914, "eval_loss": 3.27937650680542, "eval_runtime": 27.6023, "eval_samples_per_second": 161.653, "eval_steps_per_second": 20.216, "step": 78000 }, { "epoch": 1.31194117155511, "grad_norm": 9.44093132019043, "learning_rate": 2.822127873523086e-05, "loss": 2.9562, "step": 78500 }, { "epoch": 1.3202974847497284, "grad_norm": 13.62260627746582, "learning_rate": 2.8082176475517447e-05, "loss": 2.982, "step": 79000 }, { "epoch": 1.3202974847497284, "eval_loss": 3.2890851497650146, "eval_runtime": 27.5678, "eval_samples_per_second": 161.856, "eval_steps_per_second": 20.241, "step": 79000 }, { "epoch": 1.328653797944347, "grad_norm": 12.078137397766113, "learning_rate": 2.7942515572190725e-05, "loss": 2.9453, "step": 79500 }, { "epoch": 1.3370101111389654, "grad_norm": 11.467178344726562, "learning_rate": 2.7802854668863997e-05, "loss": 3.0008, "step": 80000 }, { "epoch": 1.3370101111389654, "eval_loss": 3.2852883338928223, "eval_runtime": 27.5861, "eval_samples_per_second": 161.748, "eval_steps_per_second": 20.228, "step": 80000 }, { "epoch": 1.345366424333584, "grad_norm": 14.292551040649414, "learning_rate": 2.7663473087343933e-05, "loss": 2.9664, "step": 80500 }, { "epoch": 1.3537227375282026, "grad_norm": 13.714376449584961, "learning_rate": 2.7523812184017205e-05, "loss": 2.9396, "step": 81000 }, { "epoch": 1.3537227375282026, "eval_loss": 3.2859437465667725, "eval_runtime": 27.6096, "eval_samples_per_second": 161.61, "eval_steps_per_second": 20.21, "step": 81000 }, { "epoch": 1.362079050722821, "grad_norm": 12.142716407775879, "learning_rate": 2.7384151280690483e-05, "loss": 2.9775, "step": 81500 }, { "epoch": 1.3704353639174396, "grad_norm": 11.3803071975708, "learning_rate": 2.724449037736376e-05, "loss": 2.9458, "step": 82000 }, { "epoch": 1.3704353639174396, "eval_loss": 3.278106689453125, "eval_runtime": 27.5893, "eval_samples_per_second": 161.73, "eval_steps_per_second": 20.225, "step": 82000 }, { "epoch": 1.378791677112058, "grad_norm": 16.39805030822754, "learning_rate": 2.710482947403704e-05, "loss": 3.0504, "step": 82500 }, { "epoch": 1.3871479903066768, "grad_norm": 13.994576454162598, "learning_rate": 2.6965168570710315e-05, "loss": 2.9656, "step": 83000 }, { "epoch": 1.3871479903066768, "eval_loss": 3.278665781021118, "eval_runtime": 27.5347, "eval_samples_per_second": 162.05, "eval_steps_per_second": 20.265, "step": 83000 }, { "epoch": 1.3955043035012953, "grad_norm": 11.802352905273438, "learning_rate": 2.6825507667383593e-05, "loss": 2.9786, "step": 83500 }, { "epoch": 1.4038606166959138, "grad_norm": 13.618844985961914, "learning_rate": 2.668584676405687e-05, "loss": 3.0007, "step": 84000 }, { "epoch": 1.4038606166959138, "eval_loss": 3.2725257873535156, "eval_runtime": 27.5621, "eval_samples_per_second": 161.889, "eval_steps_per_second": 20.245, "step": 84000 }, { "epoch": 1.4122169298905323, "grad_norm": 9.817100524902344, "learning_rate": 2.654618586073015e-05, "loss": 2.9268, "step": 84500 }, { "epoch": 1.4205732430851508, "grad_norm": 16.49465560913086, "learning_rate": 2.640652495740343e-05, "loss": 2.984, "step": 85000 }, { "epoch": 1.4205732430851508, "eval_loss": 3.278170108795166, "eval_runtime": 27.5927, "eval_samples_per_second": 161.71, "eval_steps_per_second": 20.223, "step": 85000 }, { "epoch": 1.4289295562797695, "grad_norm": 17.29984474182129, "learning_rate": 2.62668640540767e-05, "loss": 2.9955, "step": 85500 }, { "epoch": 1.437285869474388, "grad_norm": 12.310997009277344, "learning_rate": 2.612720315074998e-05, "loss": 2.9769, "step": 86000 }, { "epoch": 1.437285869474388, "eval_loss": 3.2687652111053467, "eval_runtime": 27.5431, "eval_samples_per_second": 162.0, "eval_steps_per_second": 20.259, "step": 86000 }, { "epoch": 1.4456421826690065, "grad_norm": 14.744447708129883, "learning_rate": 2.5987542247423257e-05, "loss": 2.966, "step": 86500 }, { "epoch": 1.453998495863625, "grad_norm": 10.83408260345459, "learning_rate": 2.5847881344096535e-05, "loss": 2.9281, "step": 87000 }, { "epoch": 1.453998495863625, "eval_loss": 3.260927677154541, "eval_runtime": 27.5537, "eval_samples_per_second": 161.938, "eval_steps_per_second": 20.251, "step": 87000 }, { "epoch": 1.4623548090582434, "grad_norm": 14.912446975708008, "learning_rate": 2.5708220440769813e-05, "loss": 2.964, "step": 87500 }, { "epoch": 1.4707111222528622, "grad_norm": 16.433135986328125, "learning_rate": 2.5568838859249743e-05, "loss": 2.9903, "step": 88000 }, { "epoch": 1.4707111222528622, "eval_loss": 3.2638683319091797, "eval_runtime": 27.5854, "eval_samples_per_second": 161.752, "eval_steps_per_second": 20.228, "step": 88000 }, { "epoch": 1.4790674354474806, "grad_norm": 10.865525245666504, "learning_rate": 2.542917795592302e-05, "loss": 2.9782, "step": 88500 }, { "epoch": 1.4874237486420991, "grad_norm": 18.059494018554688, "learning_rate": 2.5289517052596296e-05, "loss": 2.9746, "step": 89000 }, { "epoch": 1.4874237486420991, "eval_loss": 3.2576780319213867, "eval_runtime": 27.6301, "eval_samples_per_second": 161.491, "eval_steps_per_second": 20.195, "step": 89000 }, { "epoch": 1.4957800618367176, "grad_norm": 14.338726997375488, "learning_rate": 2.5149856149269575e-05, "loss": 2.9746, "step": 89500 }, { "epoch": 1.5041363750313361, "grad_norm": 16.35688018798828, "learning_rate": 2.5010195245942853e-05, "loss": 2.9235, "step": 90000 }, { "epoch": 1.5041363750313361, "eval_loss": 3.2603578567504883, "eval_runtime": 27.5275, "eval_samples_per_second": 162.093, "eval_steps_per_second": 20.271, "step": 90000 }, { "epoch": 1.5124926882259548, "grad_norm": 19.649658203125, "learning_rate": 2.4870534342616128e-05, "loss": 2.892, "step": 90500 }, { "epoch": 1.520849001420573, "grad_norm": 22.463607788085938, "learning_rate": 2.4730873439289406e-05, "loss": 2.9464, "step": 91000 }, { "epoch": 1.520849001420573, "eval_loss": 3.255012273788452, "eval_runtime": 27.5843, "eval_samples_per_second": 161.759, "eval_steps_per_second": 20.229, "step": 91000 }, { "epoch": 1.5292053146151918, "grad_norm": 11.892714500427246, "learning_rate": 2.4591212535962685e-05, "loss": 2.9404, "step": 91500 }, { "epoch": 1.5375616278098103, "grad_norm": 13.547897338867188, "learning_rate": 2.445155163263596e-05, "loss": 2.9935, "step": 92000 }, { "epoch": 1.5375616278098103, "eval_loss": 3.2467143535614014, "eval_runtime": 27.5751, "eval_samples_per_second": 161.813, "eval_steps_per_second": 20.236, "step": 92000 }, { "epoch": 1.5459179410044288, "grad_norm": 15.99018383026123, "learning_rate": 2.4311890729309238e-05, "loss": 2.9983, "step": 92500 }, { "epoch": 1.5542742541990475, "grad_norm": 10.513391494750977, "learning_rate": 2.4172229825982516e-05, "loss": 2.979, "step": 93000 }, { "epoch": 1.5542742541990475, "eval_loss": 3.2534940242767334, "eval_runtime": 27.6008, "eval_samples_per_second": 161.662, "eval_steps_per_second": 20.217, "step": 93000 }, { "epoch": 1.5626305673936658, "grad_norm": 14.598124504089355, "learning_rate": 2.4032568922655795e-05, "loss": 2.9401, "step": 93500 }, { "epoch": 1.5709868805882845, "grad_norm": 11.219178199768066, "learning_rate": 2.3893187341135724e-05, "loss": 2.9333, "step": 94000 }, { "epoch": 1.5709868805882845, "eval_loss": 3.2531471252441406, "eval_runtime": 27.5659, "eval_samples_per_second": 161.867, "eval_steps_per_second": 20.242, "step": 94000 }, { "epoch": 1.579343193782903, "grad_norm": 13.708407402038574, "learning_rate": 2.3753805759615654e-05, "loss": 2.9284, "step": 94500 }, { "epoch": 1.5876995069775215, "grad_norm": 15.64401912689209, "learning_rate": 2.3614144856288932e-05, "loss": 2.9355, "step": 95000 }, { "epoch": 1.5876995069775215, "eval_loss": 3.247119665145874, "eval_runtime": 27.5812, "eval_samples_per_second": 161.777, "eval_steps_per_second": 20.231, "step": 95000 }, { "epoch": 1.5960558201721402, "grad_norm": 12.710307121276855, "learning_rate": 2.347448395296221e-05, "loss": 2.9451, "step": 95500 }, { "epoch": 1.6044121333667585, "grad_norm": 12.77171516418457, "learning_rate": 2.333482304963549e-05, "loss": 2.904, "step": 96000 }, { "epoch": 1.6044121333667585, "eval_loss": 3.2489845752716064, "eval_runtime": 27.6157, "eval_samples_per_second": 161.575, "eval_steps_per_second": 20.206, "step": 96000 }, { "epoch": 1.6127684465613772, "grad_norm": 12.342710494995117, "learning_rate": 2.3195162146308764e-05, "loss": 2.9116, "step": 96500 }, { "epoch": 1.6211247597559957, "grad_norm": 12.343132019042969, "learning_rate": 2.3055501242982042e-05, "loss": 2.9464, "step": 97000 }, { "epoch": 1.6211247597559957, "eval_loss": 3.2335522174835205, "eval_runtime": 27.5638, "eval_samples_per_second": 161.879, "eval_steps_per_second": 20.244, "step": 97000 }, { "epoch": 1.6294810729506142, "grad_norm": 14.988670349121094, "learning_rate": 2.291611966146197e-05, "loss": 2.9173, "step": 97500 }, { "epoch": 1.6378373861452329, "grad_norm": 12.14406967163086, "learning_rate": 2.277645875813525e-05, "loss": 2.917, "step": 98000 }, { "epoch": 1.6378373861452329, "eval_loss": 3.240186929702759, "eval_runtime": 27.7378, "eval_samples_per_second": 160.864, "eval_steps_per_second": 20.117, "step": 98000 }, { "epoch": 1.6461936993398512, "grad_norm": 13.880926132202148, "learning_rate": 2.2636797854808525e-05, "loss": 2.9146, "step": 98500 }, { "epoch": 1.6545500125344699, "grad_norm": 7.802238941192627, "learning_rate": 2.2497136951481803e-05, "loss": 2.9218, "step": 99000 }, { "epoch": 1.6545500125344699, "eval_loss": 3.2389557361602783, "eval_runtime": 27.6304, "eval_samples_per_second": 161.489, "eval_steps_per_second": 20.195, "step": 99000 }, { "epoch": 1.6629063257290884, "grad_norm": 18.44457244873047, "learning_rate": 2.235747604815508e-05, "loss": 2.9043, "step": 99500 }, { "epoch": 1.6712626389237069, "grad_norm": 10.393033027648926, "learning_rate": 2.2217815144828357e-05, "loss": 2.9677, "step": 100000 }, { "epoch": 1.6712626389237069, "eval_loss": 3.2266006469726562, "eval_runtime": 27.5938, "eval_samples_per_second": 161.703, "eval_steps_per_second": 20.222, "step": 100000 }, { "epoch": 1.6796189521183253, "grad_norm": 7.137568473815918, "learning_rate": 2.2078154241501635e-05, "loss": 2.9498, "step": 100500 }, { "epoch": 1.6879752653129438, "grad_norm": 9.725958824157715, "learning_rate": 2.1938772659981565e-05, "loss": 2.8844, "step": 101000 }, { "epoch": 1.6879752653129438, "eval_loss": 3.223768949508667, "eval_runtime": 27.5732, "eval_samples_per_second": 161.824, "eval_steps_per_second": 20.237, "step": 101000 }, { "epoch": 1.6963315785075626, "grad_norm": 15.254230499267578, "learning_rate": 2.1799111756654843e-05, "loss": 2.8841, "step": 101500 }, { "epoch": 1.704687891702181, "grad_norm": 20.192659378051758, "learning_rate": 2.165945085332812e-05, "loss": 2.9283, "step": 102000 }, { "epoch": 1.704687891702181, "eval_loss": 3.2226974964141846, "eval_runtime": 27.58, "eval_samples_per_second": 161.784, "eval_steps_per_second": 20.232, "step": 102000 }, { "epoch": 1.7130442048967995, "grad_norm": 14.292362213134766, "learning_rate": 2.15197899500014e-05, "loss": 2.9358, "step": 102500 }, { "epoch": 1.721400518091418, "grad_norm": 9.396713256835938, "learning_rate": 2.1380129046674675e-05, "loss": 2.9472, "step": 103000 }, { "epoch": 1.721400518091418, "eval_loss": 3.224209785461426, "eval_runtime": 27.6671, "eval_samples_per_second": 161.274, "eval_steps_per_second": 20.168, "step": 103000 }, { "epoch": 1.7297568312860365, "grad_norm": 10.828228950500488, "learning_rate": 2.1240468143347953e-05, "loss": 2.9152, "step": 103500 }, { "epoch": 1.7381131444806552, "grad_norm": 13.493616104125977, "learning_rate": 2.1100807240021228e-05, "loss": 2.9518, "step": 104000 }, { "epoch": 1.7381131444806552, "eval_loss": 3.2257561683654785, "eval_runtime": 27.5631, "eval_samples_per_second": 161.883, "eval_steps_per_second": 20.244, "step": 104000 }, { "epoch": 1.7464694576752735, "grad_norm": 11.142574310302734, "learning_rate": 2.0961146336694506e-05, "loss": 2.9459, "step": 104500 }, { "epoch": 1.7548257708698922, "grad_norm": 10.669454574584961, "learning_rate": 2.082176475517444e-05, "loss": 2.9545, "step": 105000 }, { "epoch": 1.7548257708698922, "eval_loss": 3.2120673656463623, "eval_runtime": 27.5868, "eval_samples_per_second": 161.744, "eval_steps_per_second": 20.227, "step": 105000 }, { "epoch": 1.7631820840645107, "grad_norm": 10.605733871459961, "learning_rate": 2.0682103851847714e-05, "loss": 2.9228, "step": 105500 }, { "epoch": 1.7715383972591292, "grad_norm": 13.702558517456055, "learning_rate": 2.0542442948520993e-05, "loss": 2.9137, "step": 106000 }, { "epoch": 1.7715383972591292, "eval_loss": 3.218060255050659, "eval_runtime": 27.5869, "eval_samples_per_second": 161.744, "eval_steps_per_second": 20.227, "step": 106000 }, { "epoch": 1.779894710453748, "grad_norm": 18.355859756469727, "learning_rate": 2.0402782045194268e-05, "loss": 2.885, "step": 106500 }, { "epoch": 1.7882510236483662, "grad_norm": 12.07524299621582, "learning_rate": 2.0263679785480855e-05, "loss": 2.9016, "step": 107000 }, { "epoch": 1.7882510236483662, "eval_loss": 3.2088520526885986, "eval_runtime": 27.6024, "eval_samples_per_second": 161.653, "eval_steps_per_second": 20.216, "step": 107000 }, { "epoch": 1.796607336842985, "grad_norm": 11.443526268005371, "learning_rate": 2.0124298203960785e-05, "loss": 2.942, "step": 107500 }, { "epoch": 1.8049636500376034, "grad_norm": 8.035077095031738, "learning_rate": 1.9984637300634063e-05, "loss": 2.9247, "step": 108000 }, { "epoch": 1.8049636500376034, "eval_loss": 3.211854934692383, "eval_runtime": 27.5517, "eval_samples_per_second": 161.95, "eval_steps_per_second": 20.253, "step": 108000 }, { "epoch": 1.8133199632322219, "grad_norm": 12.948112487792969, "learning_rate": 1.9844976397307338e-05, "loss": 2.9112, "step": 108500 }, { "epoch": 1.8216762764268406, "grad_norm": 15.308154106140137, "learning_rate": 1.9705315493980616e-05, "loss": 2.9185, "step": 109000 }, { "epoch": 1.8216762764268406, "eval_loss": 3.1997740268707275, "eval_runtime": 27.6334, "eval_samples_per_second": 161.471, "eval_steps_per_second": 20.193, "step": 109000 }, { "epoch": 1.8300325896214589, "grad_norm": 10.738883018493652, "learning_rate": 1.956565459065389e-05, "loss": 2.8789, "step": 109500 }, { "epoch": 1.8383889028160776, "grad_norm": 13.379829406738281, "learning_rate": 1.942599368732717e-05, "loss": 2.9005, "step": 110000 }, { "epoch": 1.8383889028160776, "eval_loss": 3.202975273132324, "eval_runtime": 27.6255, "eval_samples_per_second": 161.517, "eval_steps_per_second": 20.199, "step": 110000 }, { "epoch": 1.846745216010696, "grad_norm": 9.87146282196045, "learning_rate": 1.9286332784000448e-05, "loss": 2.8856, "step": 110500 }, { "epoch": 1.8551015292053146, "grad_norm": 12.577339172363281, "learning_rate": 1.9146671880673726e-05, "loss": 2.9502, "step": 111000 }, { "epoch": 1.8551015292053146, "eval_loss": 3.1968445777893066, "eval_runtime": 27.5448, "eval_samples_per_second": 161.991, "eval_steps_per_second": 20.258, "step": 111000 }, { "epoch": 1.8634578423999333, "grad_norm": 12.57132625579834, "learning_rate": 1.9007010977347005e-05, "loss": 2.8951, "step": 111500 }, { "epoch": 1.8718141555945516, "grad_norm": 14.708492279052734, "learning_rate": 1.886735007402028e-05, "loss": 2.9093, "step": 112000 }, { "epoch": 1.8718141555945516, "eval_loss": 3.1939940452575684, "eval_runtime": 27.5701, "eval_samples_per_second": 161.842, "eval_steps_per_second": 20.239, "step": 112000 }, { "epoch": 1.8801704687891703, "grad_norm": 12.688665390014648, "learning_rate": 1.8727689170693558e-05, "loss": 2.885, "step": 112500 }, { "epoch": 1.8885267819837888, "grad_norm": 11.511554718017578, "learning_rate": 1.8588028267366833e-05, "loss": 2.8351, "step": 113000 }, { "epoch": 1.8885267819837888, "eval_loss": 3.1979987621307373, "eval_runtime": 27.6058, "eval_samples_per_second": 161.633, "eval_steps_per_second": 20.213, "step": 113000 }, { "epoch": 1.8968830951784073, "grad_norm": 7.7706708908081055, "learning_rate": 1.844892600765342e-05, "loss": 2.904, "step": 113500 }, { "epoch": 1.905239408373026, "grad_norm": 12.276754379272461, "learning_rate": 1.8309265104326695e-05, "loss": 2.8785, "step": 114000 }, { "epoch": 1.905239408373026, "eval_loss": 3.20162296295166, "eval_runtime": 27.5895, "eval_samples_per_second": 161.728, "eval_steps_per_second": 20.225, "step": 114000 }, { "epoch": 1.9135957215676442, "grad_norm": 11.900626182556152, "learning_rate": 1.8169604200999974e-05, "loss": 2.8922, "step": 114500 }, { "epoch": 1.921952034762263, "grad_norm": 16.15927505493164, "learning_rate": 1.802994329767325e-05, "loss": 2.8341, "step": 115000 }, { "epoch": 1.921952034762263, "eval_loss": 3.192532539367676, "eval_runtime": 27.5532, "eval_samples_per_second": 161.941, "eval_steps_per_second": 20.252, "step": 115000 }, { "epoch": 1.9303083479568814, "grad_norm": 7.972958087921143, "learning_rate": 1.7890282394346527e-05, "loss": 2.8838, "step": 115500 }, { "epoch": 1.9386646611515, "grad_norm": 10.199915885925293, "learning_rate": 1.775118013463311e-05, "loss": 2.8599, "step": 116000 }, { "epoch": 1.9386646611515, "eval_loss": 3.185673475265503, "eval_runtime": 27.8833, "eval_samples_per_second": 160.024, "eval_steps_per_second": 20.012, "step": 116000 }, { "epoch": 1.9470209743461186, "grad_norm": 12.25405216217041, "learning_rate": 1.761151923130639e-05, "loss": 2.8582, "step": 116500 }, { "epoch": 1.955377287540737, "grad_norm": 11.32104778289795, "learning_rate": 1.7471858327979664e-05, "loss": 2.9085, "step": 117000 }, { "epoch": 1.955377287540737, "eval_loss": 3.1832330226898193, "eval_runtime": 27.611, "eval_samples_per_second": 161.602, "eval_steps_per_second": 20.209, "step": 117000 }, { "epoch": 1.9637336007353556, "grad_norm": 12.988907814025879, "learning_rate": 1.7332197424652943e-05, "loss": 2.8994, "step": 117500 }, { "epoch": 1.9720899139299741, "grad_norm": 13.372990608215332, "learning_rate": 1.719253652132622e-05, "loss": 2.8882, "step": 118000 }, { "epoch": 1.9720899139299741, "eval_loss": 3.1832265853881836, "eval_runtime": 27.5792, "eval_samples_per_second": 161.789, "eval_steps_per_second": 20.233, "step": 118000 }, { "epoch": 1.9804462271245926, "grad_norm": 13.150843620300293, "learning_rate": 1.7052875617999496e-05, "loss": 2.8809, "step": 118500 }, { "epoch": 1.988802540319211, "grad_norm": 11.465882301330566, "learning_rate": 1.6913214714672774e-05, "loss": 2.8083, "step": 119000 }, { "epoch": 1.988802540319211, "eval_loss": 3.1877200603485107, "eval_runtime": 27.5585, "eval_samples_per_second": 161.91, "eval_steps_per_second": 20.248, "step": 119000 }, { "epoch": 1.9971588535138296, "grad_norm": 15.985895156860352, "learning_rate": 1.6773553811346053e-05, "loss": 2.8512, "step": 119500 }, { "epoch": 2.0055151667084483, "grad_norm": 20.66619110107422, "learning_rate": 1.663389290801933e-05, "loss": 2.6213, "step": 120000 }, { "epoch": 2.0055151667084483, "eval_loss": 3.257246494293213, "eval_runtime": 27.6181, "eval_samples_per_second": 161.561, "eval_steps_per_second": 20.204, "step": 120000 }, { "epoch": 2.0138714799030666, "grad_norm": 19.570964813232422, "learning_rate": 1.649451132649926e-05, "loss": 2.451, "step": 120500 }, { "epoch": 2.0222277930976853, "grad_norm": 17.980260848999023, "learning_rate": 1.635485042317254e-05, "loss": 2.4327, "step": 121000 }, { "epoch": 2.0222277930976853, "eval_loss": 3.2763614654541016, "eval_runtime": 27.6142, "eval_samples_per_second": 161.584, "eval_steps_per_second": 20.207, "step": 121000 }, { "epoch": 2.030584106292304, "grad_norm": 12.097052574157715, "learning_rate": 1.621546884165247e-05, "loss": 2.4341, "step": 121500 }, { "epoch": 2.0389404194869223, "grad_norm": 13.760506629943848, "learning_rate": 1.6075807938325747e-05, "loss": 2.4713, "step": 122000 }, { "epoch": 2.0389404194869223, "eval_loss": 3.28802490234375, "eval_runtime": 27.5322, "eval_samples_per_second": 162.065, "eval_steps_per_second": 20.267, "step": 122000 }, { "epoch": 2.047296732681541, "grad_norm": 13.664862632751465, "learning_rate": 1.5936147034999025e-05, "loss": 2.4953, "step": 122500 }, { "epoch": 2.0556530458761593, "grad_norm": 14.178253173828125, "learning_rate": 1.57964861316723e-05, "loss": 2.4641, "step": 123000 }, { "epoch": 2.0556530458761593, "eval_loss": 3.2855935096740723, "eval_runtime": 27.617, "eval_samples_per_second": 161.567, "eval_steps_per_second": 20.205, "step": 123000 }, { "epoch": 2.064009359070778, "grad_norm": 19.99590492248535, "learning_rate": 1.565682522834558e-05, "loss": 2.485, "step": 123500 }, { "epoch": 2.0723656722653967, "grad_norm": 17.964866638183594, "learning_rate": 1.5517164325018854e-05, "loss": 2.4679, "step": 124000 }, { "epoch": 2.0723656722653967, "eval_loss": 3.2831804752349854, "eval_runtime": 27.6195, "eval_samples_per_second": 161.553, "eval_steps_per_second": 20.203, "step": 124000 }, { "epoch": 2.080721985460015, "grad_norm": 16.556684494018555, "learning_rate": 1.5377503421692132e-05, "loss": 2.4853, "step": 124500 }, { "epoch": 2.0890782986546337, "grad_norm": 17.376474380493164, "learning_rate": 1.5237842518365409e-05, "loss": 2.4614, "step": 125000 }, { "epoch": 2.0890782986546337, "eval_loss": 3.2957663536071777, "eval_runtime": 27.5859, "eval_samples_per_second": 161.749, "eval_steps_per_second": 20.228, "step": 125000 }, { "epoch": 2.097434611849252, "grad_norm": 16.47422218322754, "learning_rate": 1.5098181615038687e-05, "loss": 2.5273, "step": 125500 }, { "epoch": 2.1057909250438707, "grad_norm": 20.20784568786621, "learning_rate": 1.4958800033518617e-05, "loss": 2.4934, "step": 126000 }, { "epoch": 2.1057909250438707, "eval_loss": 3.2978439331054688, "eval_runtime": 27.5756, "eval_samples_per_second": 161.81, "eval_steps_per_second": 20.235, "step": 126000 }, { "epoch": 2.1141472382384894, "grad_norm": 22.081459045410156, "learning_rate": 1.4819418451998548e-05, "loss": 2.4496, "step": 126500 }, { "epoch": 2.1225035514331076, "grad_norm": 14.315736770629883, "learning_rate": 1.4679757548671824e-05, "loss": 2.4892, "step": 127000 }, { "epoch": 2.1225035514331076, "eval_loss": 3.2925477027893066, "eval_runtime": 27.6072, "eval_samples_per_second": 161.624, "eval_steps_per_second": 20.212, "step": 127000 }, { "epoch": 2.1308598646277264, "grad_norm": 19.269027709960938, "learning_rate": 1.4540096645345103e-05, "loss": 2.4366, "step": 127500 }, { "epoch": 2.1392161778223446, "grad_norm": 13.193103790283203, "learning_rate": 1.4400435742018381e-05, "loss": 2.4774, "step": 128000 }, { "epoch": 2.1392161778223446, "eval_loss": 3.2914838790893555, "eval_runtime": 27.5768, "eval_samples_per_second": 161.803, "eval_steps_per_second": 20.234, "step": 128000 }, { "epoch": 2.1475724910169633, "grad_norm": 20.42424201965332, "learning_rate": 1.4260774838691656e-05, "loss": 2.4849, "step": 128500 }, { "epoch": 2.155928804211582, "grad_norm": 14.249687194824219, "learning_rate": 1.4121113935364934e-05, "loss": 2.4462, "step": 129000 }, { "epoch": 2.155928804211582, "eval_loss": 3.2914915084838867, "eval_runtime": 27.5308, "eval_samples_per_second": 162.073, "eval_steps_per_second": 20.268, "step": 129000 }, { "epoch": 2.1642851174062003, "grad_norm": 28.348491668701172, "learning_rate": 1.3981453032038211e-05, "loss": 2.49, "step": 129500 }, { "epoch": 2.172641430600819, "grad_norm": 17.39087677001953, "learning_rate": 1.384179212871149e-05, "loss": 2.4542, "step": 130000 }, { "epoch": 2.172641430600819, "eval_loss": 3.2924540042877197, "eval_runtime": 27.5959, "eval_samples_per_second": 161.691, "eval_steps_per_second": 20.22, "step": 130000 }, { "epoch": 2.1809977437954373, "grad_norm": 16.53498649597168, "learning_rate": 1.3702131225384768e-05, "loss": 2.4621, "step": 130500 }, { "epoch": 2.189354056990056, "grad_norm": 16.412384033203125, "learning_rate": 1.3562749643864697e-05, "loss": 2.4613, "step": 131000 }, { "epoch": 2.189354056990056, "eval_loss": 3.295562744140625, "eval_runtime": 27.578, "eval_samples_per_second": 161.796, "eval_steps_per_second": 20.234, "step": 131000 }, { "epoch": 2.1977103701846747, "grad_norm": 15.538711547851562, "learning_rate": 1.3423368062344627e-05, "loss": 2.4549, "step": 131500 }, { "epoch": 2.206066683379293, "grad_norm": 26.194726943969727, "learning_rate": 1.328398648082456e-05, "loss": 2.4887, "step": 132000 }, { "epoch": 2.206066683379293, "eval_loss": 3.28711199760437, "eval_runtime": 27.5596, "eval_samples_per_second": 161.904, "eval_steps_per_second": 20.247, "step": 132000 }, { "epoch": 2.2144229965739117, "grad_norm": 17.261341094970703, "learning_rate": 1.3144325577497835e-05, "loss": 2.4712, "step": 132500 }, { "epoch": 2.22277930976853, "grad_norm": 13.89609146118164, "learning_rate": 1.3004664674171113e-05, "loss": 2.505, "step": 133000 }, { "epoch": 2.22277930976853, "eval_loss": 3.281574249267578, "eval_runtime": 27.554, "eval_samples_per_second": 161.936, "eval_steps_per_second": 20.251, "step": 133000 }, { "epoch": 2.2311356229631487, "grad_norm": 20.80926513671875, "learning_rate": 1.2865003770844391e-05, "loss": 2.4657, "step": 133500 }, { "epoch": 2.2394919361577674, "grad_norm": 11.168779373168945, "learning_rate": 1.2725622189324321e-05, "loss": 2.4894, "step": 134000 }, { "epoch": 2.2394919361577674, "eval_loss": 3.277146100997925, "eval_runtime": 27.5992, "eval_samples_per_second": 161.671, "eval_steps_per_second": 20.218, "step": 134000 }, { "epoch": 2.2478482493523857, "grad_norm": 16.13945198059082, "learning_rate": 1.25859612859976e-05, "loss": 2.473, "step": 134500 }, { "epoch": 2.2562045625470044, "grad_norm": 15.301444053649902, "learning_rate": 1.2446300382670876e-05, "loss": 2.4731, "step": 135000 }, { "epoch": 2.2562045625470044, "eval_loss": 3.277038812637329, "eval_runtime": 27.7191, "eval_samples_per_second": 160.972, "eval_steps_per_second": 20.131, "step": 135000 }, { "epoch": 2.2645608757416227, "grad_norm": 14.259910583496094, "learning_rate": 1.2306639479344154e-05, "loss": 2.4209, "step": 135500 }, { "epoch": 2.2729171889362414, "grad_norm": 11.308433532714844, "learning_rate": 1.2166978576017431e-05, "loss": 2.469, "step": 136000 }, { "epoch": 2.2729171889362414, "eval_loss": 3.272756814956665, "eval_runtime": 27.5588, "eval_samples_per_second": 161.908, "eval_steps_per_second": 20.248, "step": 136000 }, { "epoch": 2.28127350213086, "grad_norm": 11.12240219116211, "learning_rate": 1.2027317672690708e-05, "loss": 2.4856, "step": 136500 }, { "epoch": 2.2896298153254784, "grad_norm": 20.829790115356445, "learning_rate": 1.1887656769363984e-05, "loss": 2.4522, "step": 137000 }, { "epoch": 2.2896298153254784, "eval_loss": 3.2834246158599854, "eval_runtime": 27.6412, "eval_samples_per_second": 161.426, "eval_steps_per_second": 20.187, "step": 137000 }, { "epoch": 2.297986128520097, "grad_norm": 16.781526565551758, "learning_rate": 1.1747995866037263e-05, "loss": 2.4827, "step": 137500 }, { "epoch": 2.3063424417147154, "grad_norm": 14.312143325805664, "learning_rate": 1.160833496271054e-05, "loss": 2.4823, "step": 138000 }, { "epoch": 2.3063424417147154, "eval_loss": 3.2799878120422363, "eval_runtime": 27.6358, "eval_samples_per_second": 161.457, "eval_steps_per_second": 20.191, "step": 138000 }, { "epoch": 2.314698754909334, "grad_norm": 16.167057037353516, "learning_rate": 1.1468674059383816e-05, "loss": 2.4814, "step": 138500 }, { "epoch": 2.3230550681039523, "grad_norm": 13.810924530029297, "learning_rate": 1.1329013156057094e-05, "loss": 2.4825, "step": 139000 }, { "epoch": 2.3230550681039523, "eval_loss": 3.278149127960205, "eval_runtime": 27.6409, "eval_samples_per_second": 161.428, "eval_steps_per_second": 20.187, "step": 139000 }, { "epoch": 2.331411381298571, "grad_norm": 15.337100982666016, "learning_rate": 1.1189352252730371e-05, "loss": 2.5083, "step": 139500 }, { "epoch": 2.3397676944931898, "grad_norm": 19.78384017944336, "learning_rate": 1.1049691349403648e-05, "loss": 2.4427, "step": 140000 }, { "epoch": 2.3397676944931898, "eval_loss": 3.276196002960205, "eval_runtime": 27.5881, "eval_samples_per_second": 161.737, "eval_steps_per_second": 20.226, "step": 140000 } ], "logging_steps": 500, "max_steps": 179505, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 2000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 6.977262699675648e+17, "train_batch_size": 8, "trial_name": null, "trial_params": null }