| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.030418250950570342, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006083650190114068, | |
| "eval_loss": 2.321223497390747, | |
| "eval_runtime": 23.8211, | |
| "eval_samples_per_second": 29.092, | |
| "eval_steps_per_second": 14.567, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006083650190114068, | |
| "grad_norm": 7.8340373039245605, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 8.7001, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00790874524714829, | |
| "eval_loss": 1.8213475942611694, | |
| "eval_runtime": 23.886, | |
| "eval_samples_per_second": 29.013, | |
| "eval_steps_per_second": 14.527, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.012167300380228136, | |
| "grad_norm": 5.482513904571533, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 7.2986, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01581749049429658, | |
| "eval_loss": 1.7394769191741943, | |
| "eval_runtime": 23.7962, | |
| "eval_samples_per_second": 29.122, | |
| "eval_steps_per_second": 14.582, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.018250950570342206, | |
| "grad_norm": 5.670393943786621, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 7.1029, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.023726235741444866, | |
| "eval_loss": 1.7079708576202393, | |
| "eval_runtime": 23.9546, | |
| "eval_samples_per_second": 28.93, | |
| "eval_steps_per_second": 14.486, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.024334600760456272, | |
| "grad_norm": 3.9268152713775635, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 6.8177, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.030418250950570342, | |
| "grad_norm": 7.93922758102417, | |
| "learning_rate": 0.0, | |
| "loss": 6.4832, | |
| "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": 8306538993156096.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |