| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12974375608173858, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025948751216347712, | |
| "grad_norm": 108.15426635742188, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 86.6731, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0025948751216347712, | |
| "eval_loss": 11.188285827636719, | |
| "eval_runtime": 77.1484, | |
| "eval_samples_per_second": 4.213, | |
| "eval_steps_per_second": 2.113, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0051897502432695424, | |
| "grad_norm": 107.09851837158203, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 86.5966, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007784625364904314, | |
| "grad_norm": 139.3520965576172, | |
| "learning_rate": 0.0001, | |
| "loss": 86.4316, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010379500486539085, | |
| "grad_norm": 96.97356414794922, | |
| "learning_rate": 9.99524110790929e-05, | |
| "loss": 76.9083, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.012974375608173857, | |
| "grad_norm": 59.65872573852539, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 69.4136, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015569250729808628, | |
| "grad_norm": 71.96248626708984, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 53.4392, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018164125851443398, | |
| "grad_norm": 63.12929153442383, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 43.0964, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02075900097307817, | |
| "grad_norm": 66.42699432373047, | |
| "learning_rate": 9.881480035599667e-05, | |
| "loss": 30.6773, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02335387609471294, | |
| "grad_norm": 76.70232391357422, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 18.9185, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.025948751216347713, | |
| "grad_norm": 56.63690185546875, | |
| "learning_rate": 9.768584753741134e-05, | |
| "loss": 7.7619, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.028543626337982485, | |
| "grad_norm": 59.534576416015625, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 5.8875, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.031138501459617256, | |
| "grad_norm": 53.30662155151367, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.8854, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03373337658125203, | |
| "grad_norm": 40.798580169677734, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 3.7457, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.036328251702886796, | |
| "grad_norm": 33.860923767089844, | |
| "learning_rate": 9.435054165891109e-05, | |
| "loss": 2.6321, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03892312682452157, | |
| "grad_norm": 34.119388580322266, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.8475, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04151800194615634, | |
| "grad_norm": 58.588966369628906, | |
| "learning_rate": 9.21695722906443e-05, | |
| "loss": 2.3496, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.044112877067791115, | |
| "grad_norm": 62.44844055175781, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 3.7984, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04670775218942588, | |
| "grad_norm": 31.974451065063477, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 1.5873, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04930262731106066, | |
| "grad_norm": 75.9893569946289, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.1131, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.051897502432695426, | |
| "grad_norm": 117.58428192138672, | |
| "learning_rate": 8.68638668405062e-05, | |
| "loss": 4.7777, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.054492377554330194, | |
| "grad_norm": 62.850887298583984, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.7957, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05708725267596497, | |
| "grad_norm": 26.412378311157227, | |
| "learning_rate": 8.377951038078302e-05, | |
| "loss": 0.9778, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.05968212779759974, | |
| "grad_norm": 15.020607948303223, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.5826, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06227700291923451, | |
| "grad_norm": 36.43248748779297, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 2.9239, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06487187804086929, | |
| "grad_norm": 25.667245864868164, | |
| "learning_rate": 7.86788218175523e-05, | |
| "loss": 0.7288, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06487187804086929, | |
| "eval_loss": 0.16820967197418213, | |
| "eval_runtime": 78.3083, | |
| "eval_samples_per_second": 4.15, | |
| "eval_steps_per_second": 2.082, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06746675316250406, | |
| "grad_norm": 15.00766658782959, | |
| "learning_rate": 7.68649804173412e-05, | |
| "loss": 0.4404, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07006162828413882, | |
| "grad_norm": 13.661465644836426, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.4163, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07265650340577359, | |
| "grad_norm": 19.321388244628906, | |
| "learning_rate": 7.308743066175172e-05, | |
| "loss": 0.6755, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07525137852740837, | |
| "grad_norm": 22.267826080322266, | |
| "learning_rate": 7.113091308703498e-05, | |
| "loss": 0.9669, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07784625364904314, | |
| "grad_norm": 7.646611213684082, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.3168, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08044112877067791, | |
| "grad_norm": 36.85639190673828, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.2373, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08303600389231268, | |
| "grad_norm": 10.440677642822266, | |
| "learning_rate": 6.503528997521366e-05, | |
| "loss": 1.3768, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08563087901394745, | |
| "grad_norm": 38.025997161865234, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.7102, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08822575413558223, | |
| "grad_norm": 10.360336303710938, | |
| "learning_rate": 6.0821980696905146e-05, | |
| "loss": 1.5797, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.090820629257217, | |
| "grad_norm": 19.017549514770508, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 1.0114, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09341550437885177, | |
| "grad_norm": 44.58971405029297, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 4.6878, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09601037950048653, | |
| "grad_norm": 30.749067306518555, | |
| "learning_rate": 5.435778713738292e-05, | |
| "loss": 1.1462, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09860525462212132, | |
| "grad_norm": 14.544421195983887, | |
| "learning_rate": 5.218096936826681e-05, | |
| "loss": 0.3816, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.10120012974375608, | |
| "grad_norm": 20.164777755737305, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1987, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10379500486539085, | |
| "grad_norm": 9.724178314208984, | |
| "learning_rate": 4.781903063173321e-05, | |
| "loss": 0.3555, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10638987998702562, | |
| "grad_norm": 24.466894149780273, | |
| "learning_rate": 4.564221286261709e-05, | |
| "loss": 0.6404, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.10898475510866039, | |
| "grad_norm": 7.151639938354492, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.3062, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11157963023029517, | |
| "grad_norm": 3.3515334129333496, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0569, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11417450535192994, | |
| "grad_norm": 24.75587272644043, | |
| "learning_rate": 3.917801930309486e-05, | |
| "loss": 0.9388, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.11676938047356471, | |
| "grad_norm": 7.871204376220703, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.1559, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11936425559519948, | |
| "grad_norm": 24.41685676574707, | |
| "learning_rate": 3.4964710024786354e-05, | |
| "loss": 0.894, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.12195913071683426, | |
| "grad_norm": 28.60738182067871, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 1.8821, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12455400583846903, | |
| "grad_norm": 29.086261749267578, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.6304, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1271488809601038, | |
| "grad_norm": 23.895931243896484, | |
| "learning_rate": 2.886908691296504e-05, | |
| "loss": 0.7597, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.12974375608173858, | |
| "grad_norm": 53.810604095458984, | |
| "learning_rate": 2.6912569338248315e-05, | |
| "loss": 2.9703, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12974375608173858, | |
| "eval_loss": 0.1682109385728836, | |
| "eval_runtime": 78.312, | |
| "eval_samples_per_second": 4.15, | |
| "eval_steps_per_second": 2.081, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.57394539429888e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |