| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.003182742463089028, | |
| "eval_steps": 9, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00017681902572716824, | |
| "eval_loss": 1.5049527883529663, | |
| "eval_runtime": 671.7063, | |
| "eval_samples_per_second": 14.18, | |
| "eval_steps_per_second": 1.773, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005304570771815047, | |
| "grad_norm": 17.419424057006836, | |
| "learning_rate": 3e-05, | |
| "loss": 6.0015, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010609141543630094, | |
| "grad_norm": 10.581345558166504, | |
| "learning_rate": 6e-05, | |
| "loss": 5.1541, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001591371231544514, | |
| "grad_norm": 7.764749526977539, | |
| "learning_rate": 9e-05, | |
| "loss": 4.2045, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001591371231544514, | |
| "eval_loss": 0.9556123614311218, | |
| "eval_runtime": 675.7088, | |
| "eval_samples_per_second": 14.096, | |
| "eval_steps_per_second": 1.763, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002121828308726019, | |
| "grad_norm": 7.4701151847839355, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.6782, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0026522853859075237, | |
| "grad_norm": 5.40271520614624, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.4578, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.003182742463089028, | |
| "grad_norm": 5.472065448760986, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.1924, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.003182742463089028, | |
| "eval_loss": 0.811770498752594, | |
| "eval_runtime": 675.7606, | |
| "eval_samples_per_second": 14.095, | |
| "eval_steps_per_second": 1.762, | |
| "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": 2.531269094355763e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |