| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0390625, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00390625, | |
| "grad_norm": 0.4705980718135834, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0788, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00390625, | |
| "eval_loss": 2.183541774749756, | |
| "eval_runtime": 3.7867, | |
| "eval_samples_per_second": 28.521, | |
| "eval_steps_per_second": 14.261, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0078125, | |
| "grad_norm": 0.46773648262023926, | |
| "learning_rate": 4e-05, | |
| "loss": 2.203, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.01171875, | |
| "grad_norm": 0.4773087501525879, | |
| "learning_rate": 6e-05, | |
| "loss": 2.3165, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01171875, | |
| "eval_loss": 2.1793529987335205, | |
| "eval_runtime": 3.0914, | |
| "eval_samples_per_second": 34.935, | |
| "eval_steps_per_second": 17.468, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.015625, | |
| "grad_norm": 0.46294865012168884, | |
| "learning_rate": 8e-05, | |
| "loss": 2.0049, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01953125, | |
| "grad_norm": 0.44666004180908203, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8536, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0234375, | |
| "grad_norm": 0.5266293287277222, | |
| "learning_rate": 0.00012, | |
| "loss": 1.9211, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0234375, | |
| "eval_loss": 2.135219097137451, | |
| "eval_runtime": 3.1069, | |
| "eval_samples_per_second": 34.761, | |
| "eval_steps_per_second": 17.381, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02734375, | |
| "grad_norm": 0.6763166189193726, | |
| "learning_rate": 0.00014, | |
| "loss": 2.1374, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03125, | |
| "grad_norm": 0.6197513341903687, | |
| "learning_rate": 0.00016, | |
| "loss": 1.8825, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03515625, | |
| "grad_norm": 0.6848791241645813, | |
| "learning_rate": 0.00018, | |
| "loss": 2.032, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03515625, | |
| "eval_loss": 1.9509679079055786, | |
| "eval_runtime": 3.1652, | |
| "eval_samples_per_second": 34.121, | |
| "eval_steps_per_second": 17.06, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0390625, | |
| "grad_norm": 0.6203467845916748, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7594, | |
| "step": 10 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1628733135912960.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |