| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.020178665265370467, | |
| "eval_steps": 9, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011210369591872482, | |
| "eval_loss": NaN, | |
| "eval_runtime": 363.8331, | |
| "eval_samples_per_second": 8.259, | |
| "eval_steps_per_second": 4.131, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0033631108775617445, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006726221755123489, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010089332632685234, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.75e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.010089332632685234, | |
| "eval_loss": NaN, | |
| "eval_runtime": 363.3391, | |
| "eval_samples_per_second": 8.271, | |
| "eval_steps_per_second": 4.137, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013452443510246978, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.016815554387808723, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001125, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.020178665265370467, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.000135, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.020178665265370467, | |
| "eval_loss": NaN, | |
| "eval_runtime": 363.475, | |
| "eval_samples_per_second": 8.267, | |
| "eval_steps_per_second": 4.135, | |
| "step": 18 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1488560963334963e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |