| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0120481927710843, | |
| "eval_steps": 11, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.024096385542168676, | |
| "eval_loss": 2.2396206855773926, | |
| "eval_runtime": 0.819, | |
| "eval_samples_per_second": 21.977, | |
| "eval_steps_per_second": 10.988, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.12048192771084337, | |
| "grad_norm": 2.930671453475952, | |
| "learning_rate": 5e-05, | |
| "loss": 8.4169, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24096385542168675, | |
| "grad_norm": 5.1463799476623535, | |
| "learning_rate": 0.0001, | |
| "loss": 9.684, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.26506024096385544, | |
| "eval_loss": 2.027172088623047, | |
| "eval_runtime": 0.8116, | |
| "eval_samples_per_second": 22.179, | |
| "eval_steps_per_second": 11.09, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3614457831325301, | |
| "grad_norm": 4.772299289703369, | |
| "learning_rate": 9.409606321741775e-05, | |
| "loss": 8.0285, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4819277108433735, | |
| "grad_norm": 5.042806148529053, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 8.0437, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5301204819277109, | |
| "eval_loss": 1.8765888214111328, | |
| "eval_runtime": 0.8213, | |
| "eval_samples_per_second": 21.917, | |
| "eval_steps_per_second": 10.958, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "grad_norm": 3.4582600593566895, | |
| "learning_rate": 5.490085701647805e-05, | |
| "loss": 7.4872, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7228915662650602, | |
| "grad_norm": 7.964458465576172, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 8.2409, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7951807228915663, | |
| "eval_loss": 1.8038959503173828, | |
| "eval_runtime": 0.8269, | |
| "eval_samples_per_second": 21.768, | |
| "eval_steps_per_second": 10.884, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.8433734939759037, | |
| "grad_norm": 4.781399726867676, | |
| "learning_rate": 1.134947733186315e-05, | |
| "loss": 7.6683, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 7.088382720947266, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 7.8901, | |
| "step": 40 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 11, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5974486312550400.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |