| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.028037383177570093, | |
| "eval_steps": 3, | |
| "global_step": 6, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004672897196261682, | |
| "grad_norm": 1.4127647876739502, | |
| "learning_rate": 2e-05, | |
| "loss": 2.2352, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004672897196261682, | |
| "eval_loss": 2.229696035385132, | |
| "eval_runtime": 4.1133, | |
| "eval_samples_per_second": 22.123, | |
| "eval_steps_per_second": 11.183, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009345794392523364, | |
| "grad_norm": 1.7348231077194214, | |
| "learning_rate": 4e-05, | |
| "loss": 2.4145, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.014018691588785047, | |
| "grad_norm": 1.5030430555343628, | |
| "learning_rate": 6e-05, | |
| "loss": 2.145, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.014018691588785047, | |
| "eval_loss": 2.215701103210449, | |
| "eval_runtime": 2.8769, | |
| "eval_samples_per_second": 31.631, | |
| "eval_steps_per_second": 15.989, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.018691588785046728, | |
| "grad_norm": 1.6331862211227417, | |
| "learning_rate": 8e-05, | |
| "loss": 2.2682, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02336448598130841, | |
| "grad_norm": 1.4806976318359375, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9887, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.028037383177570093, | |
| "grad_norm": 1.6370915174484253, | |
| "learning_rate": 0.00012, | |
| "loss": 2.0355, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.028037383177570093, | |
| "eval_loss": 1.9479806423187256, | |
| "eval_runtime": 2.8938, | |
| "eval_samples_per_second": 31.446, | |
| "eval_steps_per_second": 15.896, | |
| "step": 6 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1109738548887552.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |