| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0005409285246174491, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.3869962169678183e-05, | |
| "eval_loss": 0.3170296847820282, | |
| "eval_runtime": 1026.3822, | |
| "eval_samples_per_second": 29.578, | |
| "eval_steps_per_second": 14.789, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.934981084839092e-05, | |
| "grad_norm": 0.6361538171768188, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2943, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00013869962169678183, | |
| "grad_norm": 0.7165965437889099, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2509, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00013869962169678183, | |
| "eval_loss": 0.20657119154930115, | |
| "eval_runtime": 1002.4208, | |
| "eval_samples_per_second": 30.285, | |
| "eval_steps_per_second": 15.142, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00020804943254517272, | |
| "grad_norm": 0.8206524848937988, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1373, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00027739924339356366, | |
| "grad_norm": 0.30765119194984436, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.076, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00027739924339356366, | |
| "eval_loss": 0.048924922943115234, | |
| "eval_runtime": 1003.0336, | |
| "eval_samples_per_second": 30.266, | |
| "eval_steps_per_second": 15.133, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00034674905424195455, | |
| "grad_norm": 0.2791261374950409, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.042, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00041609886509034544, | |
| "grad_norm": 0.2505738139152527, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0624, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00041609886509034544, | |
| "eval_loss": 0.03364470601081848, | |
| "eval_runtime": 1002.5128, | |
| "eval_samples_per_second": 30.282, | |
| "eval_steps_per_second": 15.141, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0004854486759387364, | |
| "grad_norm": 0.33506065607070923, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0356, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1265549023641600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |