| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09861932938856016, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0039447731755424065, | |
| "grad_norm": 12.800737380981445, | |
| "learning_rate": 2e-05, | |
| "loss": 4.4657, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0039447731755424065, | |
| "eval_loss": 4.758890151977539, | |
| "eval_runtime": 22.2538, | |
| "eval_samples_per_second": 9.616, | |
| "eval_steps_per_second": 4.808, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007889546351084813, | |
| "grad_norm": 12.97591781616211, | |
| "learning_rate": 4e-05, | |
| "loss": 4.9917, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.011834319526627219, | |
| "grad_norm": 12.59117603302002, | |
| "learning_rate": 6e-05, | |
| "loss": 4.6719, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.011834319526627219, | |
| "eval_loss": 4.388483047485352, | |
| "eval_runtime": 22.2476, | |
| "eval_samples_per_second": 9.619, | |
| "eval_steps_per_second": 4.81, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.015779092702169626, | |
| "grad_norm": 13.937124252319336, | |
| "learning_rate": 8e-05, | |
| "loss": 4.4163, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01972386587771203, | |
| "grad_norm": 13.813149452209473, | |
| "learning_rate": 0.0001, | |
| "loss": 3.3764, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.023668639053254437, | |
| "grad_norm": 8.125162124633789, | |
| "learning_rate": 0.00012, | |
| "loss": 2.1404, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.023668639053254437, | |
| "eval_loss": 1.3306505680084229, | |
| "eval_runtime": 22.355, | |
| "eval_samples_per_second": 9.573, | |
| "eval_steps_per_second": 4.786, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.027613412228796843, | |
| "grad_norm": 8.103511810302734, | |
| "learning_rate": 0.00014, | |
| "loss": 1.3031, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03155818540433925, | |
| "grad_norm": 6.617693901062012, | |
| "learning_rate": 0.00016, | |
| "loss": 0.8372, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03550295857988166, | |
| "grad_norm": 3.7571351528167725, | |
| "learning_rate": 0.00018, | |
| "loss": 0.396, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03550295857988166, | |
| "eval_loss": 0.2699667811393738, | |
| "eval_runtime": 22.2641, | |
| "eval_samples_per_second": 9.612, | |
| "eval_steps_per_second": 4.806, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03944773175542406, | |
| "grad_norm": 2.737286329269409, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3586, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04339250493096647, | |
| "grad_norm": 2.9477121829986572, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.357, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.047337278106508875, | |
| "grad_norm": 3.2014811038970947, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.3266, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.047337278106508875, | |
| "eval_loss": 0.2208854705095291, | |
| "eval_runtime": 22.3445, | |
| "eval_samples_per_second": 9.577, | |
| "eval_steps_per_second": 4.789, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05128205128205128, | |
| "grad_norm": 1.2773289680480957, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.1722, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.055226824457593686, | |
| "grad_norm": 2.2797818183898926, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.1792, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05917159763313609, | |
| "grad_norm": 1.130849838256836, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.1988, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05917159763313609, | |
| "eval_loss": 0.20907557010650635, | |
| "eval_runtime": 22.2441, | |
| "eval_samples_per_second": 9.621, | |
| "eval_steps_per_second": 4.81, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0631163708086785, | |
| "grad_norm": 1.6371982097625732, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.336, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0670611439842209, | |
| "grad_norm": 1.4365739822387695, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.2788, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "grad_norm": 0.5848339796066284, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.0984, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "eval_loss": 0.21445059776306152, | |
| "eval_runtime": 22.3011, | |
| "eval_samples_per_second": 9.596, | |
| "eval_steps_per_second": 4.798, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07495069033530571, | |
| "grad_norm": 2.026401996612549, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.2734, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.07889546351084813, | |
| "grad_norm": 1.0910876989364624, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.215, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08284023668639054, | |
| "grad_norm": 1.227197527885437, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.2262, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08284023668639054, | |
| "eval_loss": 0.1881045401096344, | |
| "eval_runtime": 22.3414, | |
| "eval_samples_per_second": 9.579, | |
| "eval_steps_per_second": 4.789, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08678500986193294, | |
| "grad_norm": 0.9256035089492798, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.1988, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.09072978303747535, | |
| "grad_norm": 0.8593224883079529, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.2024, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "grad_norm": 1.7054708003997803, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.277, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "eval_loss": 0.18979641795158386, | |
| "eval_runtime": 22.2705, | |
| "eval_samples_per_second": 9.609, | |
| "eval_steps_per_second": 4.805, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09861932938856016, | |
| "grad_norm": 0.966243326663971, | |
| "learning_rate": 0.0, | |
| "loss": 0.1495, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.26729302278144e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |