| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0008898499268098436, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.779699853619687e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1107.8259, | |
| "eval_samples_per_second": 21.356, | |
| "eval_steps_per_second": 10.679, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001779699853619687, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00023136098097055932, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1107.6866, | |
| "eval_samples_per_second": 21.359, | |
| "eval_steps_per_second": 10.68, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0003559399707239374, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00046272196194111864, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1108.8985, | |
| "eval_samples_per_second": 21.336, | |
| "eval_steps_per_second": 10.668, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0005339099560859061, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0006940829429116779, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1107.2845, | |
| "eval_samples_per_second": 21.367, | |
| "eval_steps_per_second": 10.684, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0007118799414478749, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0008898499268098436, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.586645434368e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |