| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 2000, | |
| "global_step": 5352, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001868460388639761, | |
| "grad_norm": 8.915643692016602, | |
| "learning_rate": 0.0, | |
| "loss": 3.2115, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.37369207772795215, | |
| "grad_norm": 3.4466991424560547, | |
| "learning_rate": 2.994527998755732e-05, | |
| "loss": 1.907, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.37369207772795215, | |
| "eval_cosine_accuracy": 0.9618272185325623, | |
| "eval_loss": 0.6304187178611755, | |
| "eval_runtime": 27.4998, | |
| "eval_samples_per_second": 343.893, | |
| "eval_steps_per_second": 1.345, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7473841554559043, | |
| "grad_norm": 2.7122421264648438, | |
| "learning_rate": 2.8015311122967933e-05, | |
| "loss": 1.1209, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7473841554559043, | |
| "eval_cosine_accuracy": 0.9697578549385071, | |
| "eval_loss": 0.553921639919281, | |
| "eval_runtime": 25.7506, | |
| "eval_samples_per_second": 367.254, | |
| "eval_steps_per_second": 1.437, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 16056, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |