| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 30, | |
| "global_step": 450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 0.36675021052360535, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.3008, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 0.045268554240465164, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.0061, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.006114976014941931, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0042, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "eval_loss": 0.004080221988260746, | |
| "eval_runtime": 96.8324, | |
| "eval_samples_per_second": 10.327, | |
| "eval_steps_per_second": 0.258, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 0.0014940646942704916, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0039, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.0011621147859841585, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0039, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 0.005421994719654322, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0039, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "eval_loss": 0.003909846767783165, | |
| "eval_runtime": 96.8186, | |
| "eval_samples_per_second": 10.329, | |
| "eval_steps_per_second": 0.258, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3111111111111111, | |
| "grad_norm": 0.002968616783618927, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0039, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.016864212229847908, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0039, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.004826202057301998, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0039, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.0038871022406965494, | |
| "eval_runtime": 96.843, | |
| "eval_samples_per_second": 10.326, | |
| "eval_steps_per_second": 0.258, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.003356265602633357, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0039, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4888888888888889, | |
| "grad_norm": 0.0016542698722332716, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0039, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.007389770820736885, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0039, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "eval_loss": 0.0038501631934195757, | |
| "eval_runtime": 96.845, | |
| "eval_samples_per_second": 10.326, | |
| "eval_steps_per_second": 0.258, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5777777777777777, | |
| "grad_norm": 0.009115256369113922, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0038, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 0.006779392249882221, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0037, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.022685615345835686, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0033, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "eval_loss": 0.0030691924039274454, | |
| "eval_runtime": 96.8893, | |
| "eval_samples_per_second": 10.321, | |
| "eval_steps_per_second": 0.258, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.04725850373506546, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0031, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7555555555555555, | |
| "grad_norm": 0.02934485487639904, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.0031, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.020546529442071915, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0029, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.0029199642594903708, | |
| "eval_runtime": 96.8327, | |
| "eval_samples_per_second": 10.327, | |
| "eval_steps_per_second": 0.258, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8444444444444444, | |
| "grad_norm": 0.01967918500304222, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0029, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.013359949924051762, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0028, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "grad_norm": 0.03619702160358429, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.0028, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "eval_loss": 0.002507995581254363, | |
| "eval_runtime": 96.8365, | |
| "eval_samples_per_second": 10.327, | |
| "eval_steps_per_second": 0.258, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 0.015583088621497154, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0026, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0222222222222221, | |
| "grad_norm": 0.02111578918993473, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.0028, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.014296974055469036, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0025, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "eval_loss": 0.0024458724074065685, | |
| "eval_runtime": 96.9524, | |
| "eval_samples_per_second": 10.314, | |
| "eval_steps_per_second": 0.258, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.036929819732904434, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0025, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1555555555555554, | |
| "grad_norm": 0.01881023496389389, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.0026, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.02219155989587307, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0024, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.002419873373582959, | |
| "eval_runtime": 96.9311, | |
| "eval_samples_per_second": 10.317, | |
| "eval_steps_per_second": 0.258, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2444444444444445, | |
| "grad_norm": 0.015554447658360004, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0026, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2888888888888888, | |
| "grad_norm": 0.014902675524353981, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.0026, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.023787055164575577, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0027, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "eval_loss": 0.0023624880705028772, | |
| "eval_runtime": 96.9344, | |
| "eval_samples_per_second": 10.316, | |
| "eval_steps_per_second": 0.258, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3777777777777778, | |
| "grad_norm": 0.024232730269432068, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0024, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.030378742143511772, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0025, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4666666666666668, | |
| "grad_norm": 0.022359298542141914, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 0.0025, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.4666666666666668, | |
| "eval_loss": 0.0022804364562034607, | |
| "eval_runtime": 96.9112, | |
| "eval_samples_per_second": 10.319, | |
| "eval_steps_per_second": 0.258, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.511111111111111, | |
| "grad_norm": 0.01789114437997341, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0024, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "grad_norm": 0.019261475652456284, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0027, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.021873965859413147, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0023, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.002201332012191415, | |
| "eval_runtime": 96.9199, | |
| "eval_samples_per_second": 10.318, | |
| "eval_steps_per_second": 0.258, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6444444444444444, | |
| "grad_norm": 0.0286718662828207, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0025, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6888888888888889, | |
| "grad_norm": 0.019971389323472977, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 0.0022, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.7333333333333334, | |
| "grad_norm": 0.02623485028743744, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.0023, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7333333333333334, | |
| "eval_loss": 0.002075706608593464, | |
| "eval_runtime": 96.9432, | |
| "eval_samples_per_second": 10.315, | |
| "eval_steps_per_second": 0.258, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.021184084936976433, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.002, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8222222222222222, | |
| "grad_norm": 0.02332981303334236, | |
| "learning_rate": 1.9369152030840556e-06, | |
| "loss": 0.002, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.8666666666666667, | |
| "grad_norm": 0.023150313645601273, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.002, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.8666666666666667, | |
| "eval_loss": 0.0020291407126933336, | |
| "eval_runtime": 96.9072, | |
| "eval_samples_per_second": 10.319, | |
| "eval_steps_per_second": 0.258, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.911111111111111, | |
| "grad_norm": 0.03483868017792702, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.002, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.9555555555555557, | |
| "grad_norm": 0.03138676658272743, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.0025, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.028902092948555946, | |
| "learning_rate": 0.0, | |
| "loss": 0.0022, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.0020088243763893843, | |
| "eval_runtime": 96.9299, | |
| "eval_samples_per_second": 10.317, | |
| "eval_steps_per_second": 0.258, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 450, | |
| "total_flos": 1.4721820638419354e+18, | |
| "train_loss": 0.009612742554810313, | |
| "train_runtime": 7000.1011, | |
| "train_samples_per_second": 2.571, | |
| "train_steps_per_second": 0.064 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4721820638419354e+18, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |