| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.31217481789802287, | |
| "eval_steps": 20, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04162330905306972, | |
| "grad_norm": 20.466537475585938, | |
| "learning_rate": 4.957081545064378e-05, | |
| "loss": 8.8593, | |
| "mean_token_accuracy": 0.6828776024281978, | |
| "num_tokens": 161548.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04162330905306972, | |
| "eval_loss": 1.1315333843231201, | |
| "eval_mean_token_accuracy": 0.8237503719329834, | |
| "eval_num_tokens": 161548.0, | |
| "eval_runtime": 34.7122, | |
| "eval_samples_per_second": 2.881, | |
| "eval_steps_per_second": 0.72, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08324661810613944, | |
| "grad_norm": 2.744436025619507, | |
| "learning_rate": 4.742489270386266e-05, | |
| "loss": 1.8553, | |
| "mean_token_accuracy": 0.9392111219465733, | |
| "num_tokens": 323168.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08324661810613944, | |
| "eval_loss": 0.16864576935768127, | |
| "eval_mean_token_accuracy": 0.9810383248329163, | |
| "eval_num_tokens": 323168.0, | |
| "eval_runtime": 34.3084, | |
| "eval_samples_per_second": 2.915, | |
| "eval_steps_per_second": 0.729, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12486992715920915, | |
| "grad_norm": 1.6022675037384033, | |
| "learning_rate": 4.527896995708155e-05, | |
| "loss": 0.3903, | |
| "mean_token_accuracy": 0.9888123281300067, | |
| "num_tokens": 484780.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12486992715920915, | |
| "eval_loss": 0.03530249744653702, | |
| "eval_mean_token_accuracy": 0.9926152086257934, | |
| "eval_num_tokens": 484780.0, | |
| "eval_runtime": 34.7758, | |
| "eval_samples_per_second": 2.876, | |
| "eval_steps_per_second": 0.719, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16649323621227888, | |
| "grad_norm": 1.2044650316238403, | |
| "learning_rate": 4.313304721030043e-05, | |
| "loss": 0.0664, | |
| "mean_token_accuracy": 0.9934882044792175, | |
| "num_tokens": 646431.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16649323621227888, | |
| "eval_loss": 0.012011010199785233, | |
| "eval_mean_token_accuracy": 0.9936227130889893, | |
| "eval_num_tokens": 646431.0, | |
| "eval_runtime": 34.3329, | |
| "eval_samples_per_second": 2.913, | |
| "eval_steps_per_second": 0.728, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2081165452653486, | |
| "grad_norm": 0.4138229191303253, | |
| "learning_rate": 4.098712446351932e-05, | |
| "loss": 0.0513, | |
| "mean_token_accuracy": 0.994041533768177, | |
| "num_tokens": 808054.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2081165452653486, | |
| "eval_loss": 0.011702906340360641, | |
| "eval_mean_token_accuracy": 0.9942703366279602, | |
| "eval_num_tokens": 808054.0, | |
| "eval_runtime": 34.3225, | |
| "eval_samples_per_second": 2.914, | |
| "eval_steps_per_second": 0.728, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2497398543184183, | |
| "grad_norm": 0.8489187955856323, | |
| "learning_rate": 3.88412017167382e-05, | |
| "loss": 0.0416, | |
| "mean_token_accuracy": 0.9943524189293385, | |
| "num_tokens": 969623.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2497398543184183, | |
| "eval_loss": 0.010666043497622013, | |
| "eval_mean_token_accuracy": 0.9943506979942321, | |
| "eval_num_tokens": 969623.0, | |
| "eval_runtime": 34.3761, | |
| "eval_samples_per_second": 2.909, | |
| "eval_steps_per_second": 0.727, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.29136316337148804, | |
| "grad_norm": 0.49626249074935913, | |
| "learning_rate": 3.669527896995708e-05, | |
| "loss": 0.0384, | |
| "mean_token_accuracy": 0.9943184182047844, | |
| "num_tokens": 1131222.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.29136316337148804, | |
| "eval_loss": 0.009807135909795761, | |
| "eval_mean_token_accuracy": 0.9946736145019531, | |
| "eval_num_tokens": 1131222.0, | |
| "eval_runtime": 34.34, | |
| "eval_samples_per_second": 2.912, | |
| "eval_steps_per_second": 0.728, | |
| "step": 140 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 481, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 3.16484363381472e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |