| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.004225560414950032, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.451120829900065e-05, | |
| "eval_loss": 5.234640121459961, | |
| "eval_runtime": 80.2984, | |
| "eval_samples_per_second": 62.056, | |
| "eval_steps_per_second": 31.034, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008451120829900065, | |
| "grad_norm": 1.3723881244659424, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 4.2304, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0010986457078870084, | |
| "eval_loss": 3.989668369293213, | |
| "eval_runtime": 79.9778, | |
| "eval_samples_per_second": 62.305, | |
| "eval_steps_per_second": 31.159, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.001690224165980013, | |
| "grad_norm": 1.000081181526184, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 3.4734, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.002197291415774017, | |
| "eval_loss": 3.8006229400634766, | |
| "eval_runtime": 80.0245, | |
| "eval_samples_per_second": 62.268, | |
| "eval_steps_per_second": 31.14, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0025353362489700195, | |
| "grad_norm": 0.6566107273101807, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 3.411, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0032959371236610255, | |
| "eval_loss": 3.748792886734009, | |
| "eval_runtime": 80.9379, | |
| "eval_samples_per_second": 61.566, | |
| "eval_steps_per_second": 30.789, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.003380448331960026, | |
| "grad_norm": 0.7871419787406921, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 3.3147, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.004225560414950032, | |
| "grad_norm": 2.028604507446289, | |
| "learning_rate": 0.0, | |
| "loss": 3.6075, | |
| "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": 464179005751296.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |