| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 1908, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07861635220125786, | |
| "grad_norm": 0.30994901061058044, | |
| "learning_rate": 0.00013089005235602096, | |
| "loss": 4.0802, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07861635220125786, | |
| "eval_loss": 3.3922524452209473, | |
| "eval_runtime": 27.5047, | |
| "eval_samples_per_second": 3.636, | |
| "eval_steps_per_second": 0.473, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15723270440251572, | |
| "grad_norm": 0.1917889267206192, | |
| "learning_rate": 0.0002617801047120419, | |
| "loss": 2.6833, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15723270440251572, | |
| "eval_loss": 2.327249050140381, | |
| "eval_runtime": 27.3498, | |
| "eval_samples_per_second": 3.656, | |
| "eval_steps_per_second": 0.475, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2358490566037736, | |
| "grad_norm": 0.1460731029510498, | |
| "learning_rate": 0.00039267015706806284, | |
| "loss": 2.1534, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2358490566037736, | |
| "eval_loss": 2.1624627113342285, | |
| "eval_runtime": 27.2554, | |
| "eval_samples_per_second": 3.669, | |
| "eval_steps_per_second": 0.477, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.31446540880503143, | |
| "grad_norm": 0.15231014788150787, | |
| "learning_rate": 0.0004973791496796739, | |
| "loss": 2.0476, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.31446540880503143, | |
| "eval_loss": 2.100531816482544, | |
| "eval_runtime": 27.3815, | |
| "eval_samples_per_second": 3.652, | |
| "eval_steps_per_second": 0.475, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39308176100628933, | |
| "grad_norm": 0.1586989462375641, | |
| "learning_rate": 0.00048281887012230633, | |
| "loss": 1.9929, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.39308176100628933, | |
| "eval_loss": 2.0717108249664307, | |
| "eval_runtime": 27.458, | |
| "eval_samples_per_second": 3.642, | |
| "eval_steps_per_second": 0.473, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4716981132075472, | |
| "grad_norm": 0.14461246132850647, | |
| "learning_rate": 0.00046825859056493884, | |
| "loss": 1.974, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4716981132075472, | |
| "eval_loss": 2.0508835315704346, | |
| "eval_runtime": 27.4544, | |
| "eval_samples_per_second": 3.642, | |
| "eval_steps_per_second": 0.474, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.550314465408805, | |
| "grad_norm": 0.22759944200515747, | |
| "learning_rate": 0.00045369831100757136, | |
| "loss": 1.9854, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.550314465408805, | |
| "eval_loss": 2.043459892272949, | |
| "eval_runtime": 27.4852, | |
| "eval_samples_per_second": 3.638, | |
| "eval_steps_per_second": 0.473, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6289308176100629, | |
| "grad_norm": 0.1476861983537674, | |
| "learning_rate": 0.0004391380314502039, | |
| "loss": 2.0109, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6289308176100629, | |
| "eval_loss": 2.035849094390869, | |
| "eval_runtime": 27.3716, | |
| "eval_samples_per_second": 3.653, | |
| "eval_steps_per_second": 0.475, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7075471698113207, | |
| "grad_norm": 0.16703520715236664, | |
| "learning_rate": 0.0004245777518928364, | |
| "loss": 1.9939, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7075471698113207, | |
| "eval_loss": 2.0285253524780273, | |
| "eval_runtime": 27.4183, | |
| "eval_samples_per_second": 3.647, | |
| "eval_steps_per_second": 0.474, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7861635220125787, | |
| "grad_norm": 0.14345824718475342, | |
| "learning_rate": 0.00041001747233546885, | |
| "loss": 1.9672, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7861635220125787, | |
| "eval_loss": 2.0150387287139893, | |
| "eval_runtime": 27.4063, | |
| "eval_samples_per_second": 3.649, | |
| "eval_steps_per_second": 0.474, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8647798742138365, | |
| "grad_norm": 0.15760189294815063, | |
| "learning_rate": 0.0003954571927781013, | |
| "loss": 1.9565, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8647798742138365, | |
| "eval_loss": 2.011507034301758, | |
| "eval_runtime": 27.3979, | |
| "eval_samples_per_second": 3.65, | |
| "eval_steps_per_second": 0.474, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9433962264150944, | |
| "grad_norm": 0.19006682932376862, | |
| "learning_rate": 0.0003808969132207338, | |
| "loss": 1.9706, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9433962264150944, | |
| "eval_loss": 2.006178379058838, | |
| "eval_runtime": 27.3191, | |
| "eval_samples_per_second": 3.66, | |
| "eval_steps_per_second": 0.476, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0220125786163523, | |
| "grad_norm": 0.13335242867469788, | |
| "learning_rate": 0.00036633663366336634, | |
| "loss": 1.9352, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.0220125786163523, | |
| "eval_loss": 1.9949337244033813, | |
| "eval_runtime": 27.384, | |
| "eval_samples_per_second": 3.652, | |
| "eval_steps_per_second": 0.475, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.10062893081761, | |
| "grad_norm": 0.1745985597372055, | |
| "learning_rate": 0.00035177635410599885, | |
| "loss": 1.8647, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.10062893081761, | |
| "eval_loss": 1.9930100440979004, | |
| "eval_runtime": 27.4302, | |
| "eval_samples_per_second": 3.646, | |
| "eval_steps_per_second": 0.474, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.179245283018868, | |
| "grad_norm": 0.17578236758708954, | |
| "learning_rate": 0.00033721607454863137, | |
| "loss": 1.9109, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.179245283018868, | |
| "eval_loss": 1.9840075969696045, | |
| "eval_runtime": 27.4312, | |
| "eval_samples_per_second": 3.645, | |
| "eval_steps_per_second": 0.474, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2578616352201257, | |
| "grad_norm": 0.16206932067871094, | |
| "learning_rate": 0.0003226557949912638, | |
| "loss": 1.885, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2578616352201257, | |
| "eval_loss": 1.9805808067321777, | |
| "eval_runtime": 27.4061, | |
| "eval_samples_per_second": 3.649, | |
| "eval_steps_per_second": 0.474, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3364779874213837, | |
| "grad_norm": 0.13977627456188202, | |
| "learning_rate": 0.00030809551543389634, | |
| "loss": 1.8864, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.3364779874213837, | |
| "eval_loss": 1.9878467321395874, | |
| "eval_runtime": 27.3272, | |
| "eval_samples_per_second": 3.659, | |
| "eval_steps_per_second": 0.476, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.4150943396226414, | |
| "grad_norm": 0.19199030101299286, | |
| "learning_rate": 0.00029353523587652885, | |
| "loss": 1.8931, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4150943396226414, | |
| "eval_loss": 1.9824912548065186, | |
| "eval_runtime": 27.2809, | |
| "eval_samples_per_second": 3.666, | |
| "eval_steps_per_second": 0.477, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4937106918238994, | |
| "grad_norm": 0.17020899057388306, | |
| "learning_rate": 0.0002789749563191613, | |
| "loss": 1.8599, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.4937106918238994, | |
| "eval_loss": 1.975517988204956, | |
| "eval_runtime": 27.4147, | |
| "eval_samples_per_second": 3.648, | |
| "eval_steps_per_second": 0.474, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.2071741223335266, | |
| "learning_rate": 0.00026441467676179383, | |
| "loss": 1.9019, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "eval_loss": 1.9695261716842651, | |
| "eval_runtime": 27.3628, | |
| "eval_samples_per_second": 3.655, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.650943396226415, | |
| "grad_norm": 0.18793335556983948, | |
| "learning_rate": 0.00024985439720442634, | |
| "loss": 1.9081, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.650943396226415, | |
| "eval_loss": 1.959928274154663, | |
| "eval_runtime": 27.4577, | |
| "eval_samples_per_second": 3.642, | |
| "eval_steps_per_second": 0.473, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.7295597484276728, | |
| "grad_norm": 0.2016116827726364, | |
| "learning_rate": 0.00023529411764705883, | |
| "loss": 1.8736, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.7295597484276728, | |
| "eval_loss": 1.9582524299621582, | |
| "eval_runtime": 27.3035, | |
| "eval_samples_per_second": 3.663, | |
| "eval_steps_per_second": 0.476, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.808176100628931, | |
| "grad_norm": 0.1896527260541916, | |
| "learning_rate": 0.00022073383808969132, | |
| "loss": 1.8939, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.808176100628931, | |
| "eval_loss": 1.9566823244094849, | |
| "eval_runtime": 27.4057, | |
| "eval_samples_per_second": 3.649, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.8867924528301887, | |
| "grad_norm": 0.17931561172008514, | |
| "learning_rate": 0.00020617355853232383, | |
| "loss": 1.8867, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8867924528301887, | |
| "eval_loss": 1.9534107446670532, | |
| "eval_runtime": 27.4354, | |
| "eval_samples_per_second": 3.645, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9654088050314464, | |
| "grad_norm": 0.21005648374557495, | |
| "learning_rate": 0.00019161327897495632, | |
| "loss": 1.8875, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.9654088050314464, | |
| "eval_loss": 1.947943925857544, | |
| "eval_runtime": 27.4109, | |
| "eval_samples_per_second": 3.648, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.0440251572327046, | |
| "grad_norm": 0.175890251994133, | |
| "learning_rate": 0.0001770529994175888, | |
| "loss": 1.8677, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.0440251572327046, | |
| "eval_loss": 1.949751853942871, | |
| "eval_runtime": 27.3631, | |
| "eval_samples_per_second": 3.655, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.1226415094339623, | |
| "grad_norm": 0.23929914832115173, | |
| "learning_rate": 0.00016249271986022132, | |
| "loss": 1.8152, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.1226415094339623, | |
| "eval_loss": 1.9526574611663818, | |
| "eval_runtime": 27.3536, | |
| "eval_samples_per_second": 3.656, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.20125786163522, | |
| "grad_norm": 0.25344499945640564, | |
| "learning_rate": 0.00014793244030285383, | |
| "loss": 1.8573, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.20125786163522, | |
| "eval_loss": 1.9523253440856934, | |
| "eval_runtime": 27.4028, | |
| "eval_samples_per_second": 3.649, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.279874213836478, | |
| "grad_norm": 0.23211850225925446, | |
| "learning_rate": 0.0001333721607454863, | |
| "loss": 1.8433, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.279874213836478, | |
| "eval_loss": 1.9440948963165283, | |
| "eval_runtime": 27.3179, | |
| "eval_samples_per_second": 3.661, | |
| "eval_steps_per_second": 0.476, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.358490566037736, | |
| "grad_norm": 0.23775285482406616, | |
| "learning_rate": 0.00011881188118811881, | |
| "loss": 1.828, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.358490566037736, | |
| "eval_loss": 1.945542335510254, | |
| "eval_runtime": 27.3283, | |
| "eval_samples_per_second": 3.659, | |
| "eval_steps_per_second": 0.476, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4371069182389937, | |
| "grad_norm": 0.20624059438705444, | |
| "learning_rate": 0.00010425160163075131, | |
| "loss": 1.8298, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.4371069182389937, | |
| "eval_loss": 1.9422754049301147, | |
| "eval_runtime": 27.398, | |
| "eval_samples_per_second": 3.65, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.5157232704402515, | |
| "grad_norm": 0.20326970517635345, | |
| "learning_rate": 8.969132207338381e-05, | |
| "loss": 1.8258, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.5157232704402515, | |
| "eval_loss": 1.9376040697097778, | |
| "eval_runtime": 27.3017, | |
| "eval_samples_per_second": 3.663, | |
| "eval_steps_per_second": 0.476, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.5943396226415096, | |
| "grad_norm": 0.19999054074287415, | |
| "learning_rate": 7.513104251601631e-05, | |
| "loss": 1.8314, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.5943396226415096, | |
| "eval_loss": 1.937279462814331, | |
| "eval_runtime": 27.3919, | |
| "eval_samples_per_second": 3.651, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.6729559748427674, | |
| "grad_norm": 0.17829832434654236, | |
| "learning_rate": 6.0570762958648805e-05, | |
| "loss": 1.8436, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.6729559748427674, | |
| "eval_loss": 1.93943452835083, | |
| "eval_runtime": 27.4007, | |
| "eval_samples_per_second": 3.65, | |
| "eval_steps_per_second": 0.474, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.751572327044025, | |
| "grad_norm": 0.18850775063037872, | |
| "learning_rate": 4.6010483401281306e-05, | |
| "loss": 1.8253, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.751572327044025, | |
| "eval_loss": 1.937272310256958, | |
| "eval_runtime": 27.2797, | |
| "eval_samples_per_second": 3.666, | |
| "eval_steps_per_second": 0.477, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.830188679245283, | |
| "grad_norm": 0.19603122770786285, | |
| "learning_rate": 3.145020384391381e-05, | |
| "loss": 1.8055, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.830188679245283, | |
| "eval_loss": 1.9364655017852783, | |
| "eval_runtime": 27.2904, | |
| "eval_samples_per_second": 3.664, | |
| "eval_steps_per_second": 0.476, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.908805031446541, | |
| "grad_norm": 0.22059138119220734, | |
| "learning_rate": 1.68899242865463e-05, | |
| "loss": 1.8207, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.908805031446541, | |
| "eval_loss": 1.9341883659362793, | |
| "eval_runtime": 27.2939, | |
| "eval_samples_per_second": 3.664, | |
| "eval_steps_per_second": 0.476, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.9874213836477987, | |
| "grad_norm": 0.20625567436218262, | |
| "learning_rate": 2.3296447291788002e-06, | |
| "loss": 1.8514, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.9874213836477987, | |
| "eval_loss": 1.9331157207489014, | |
| "eval_runtime": 27.3592, | |
| "eval_samples_per_second": 3.655, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1900 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1908, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 2.377867800895488e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |