| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9985964912280703, | |
| "eval_steps": 500, | |
| "global_step": 356, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 11.060110092163086, | |
| "learning_rate": 9.53723622631339e-06, | |
| "loss": 3.712, | |
| "num_input_tokens_seen": 825600, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 0.08450744301080704, | |
| "learning_rate": 8.20073735379715e-06, | |
| "loss": 0.1276, | |
| "num_input_tokens_seen": 1651200, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 0.019924940541386604, | |
| "learning_rate": 6.247734714199105e-06, | |
| "loss": 0.0644, | |
| "num_input_tokens_seen": 2476800, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1228070175438596, | |
| "grad_norm": 0.03810443729162216, | |
| "learning_rate": 4.054402092289354e-06, | |
| "loss": 0.0091, | |
| "num_input_tokens_seen": 3302400, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 0.1745239943265915, | |
| "learning_rate": 2.0432039624699016e-06, | |
| "loss": 0.0451, | |
| "num_input_tokens_seen": 4128000, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6842105263157894, | |
| "grad_norm": 0.028025347739458084, | |
| "learning_rate": 6.015233200716874e-07, | |
| "loss": 0.0004, | |
| "num_input_tokens_seen": 4953600, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.9649122807017543, | |
| "grad_norm": 0.008897515013813972, | |
| "learning_rate": 7.046663148979616e-09, | |
| "loss": 0.0069, | |
| "num_input_tokens_seen": 5779200, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.9985964912280703, | |
| "num_input_tokens_seen": 5878272, | |
| "step": 356, | |
| "total_flos": 8.486575409523917e+16, | |
| "train_loss": 0.5569543002220394, | |
| "train_runtime": 4198.4776, | |
| "train_samples_per_second": 1.358, | |
| "train_steps_per_second": 0.085 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 356, | |
| "num_input_tokens_seen": 5878272, | |
| "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": 8.486575409523917e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |