| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06565449322938038, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013130898645876078, | |
| "eval_loss": 12.693028450012207, | |
| "eval_runtime": 275.4401, | |
| "eval_samples_per_second": 18.628, | |
| "eval_steps_per_second": 2.331, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.013130898645876077, | |
| "grad_norm": 3.297091007232666, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 6.5628, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013130898645876077, | |
| "eval_loss": 0.33288654685020447, | |
| "eval_runtime": 275.5298, | |
| "eval_samples_per_second": 18.622, | |
| "eval_steps_per_second": 2.33, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.026261797291752154, | |
| "grad_norm": 1.9047911167144775, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.2901, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.026261797291752154, | |
| "eval_loss": 0.1979968100786209, | |
| "eval_runtime": 275.6902, | |
| "eval_samples_per_second": 18.611, | |
| "eval_steps_per_second": 2.329, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03939269593762823, | |
| "grad_norm": 2.357926607131958, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.21, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03939269593762823, | |
| "eval_loss": 0.1843382865190506, | |
| "eval_runtime": 275.5279, | |
| "eval_samples_per_second": 18.622, | |
| "eval_steps_per_second": 2.33, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05252359458350431, | |
| "grad_norm": 0.5868324041366577, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.1884, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05252359458350431, | |
| "eval_loss": 0.17553400993347168, | |
| "eval_runtime": 276.0685, | |
| "eval_samples_per_second": 18.586, | |
| "eval_steps_per_second": 2.326, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06565449322938038, | |
| "grad_norm": 1.1534255743026733, | |
| "learning_rate": 0.0, | |
| "loss": 0.1773, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06565449322938038, | |
| "eval_loss": 0.17201422154903412, | |
| "eval_runtime": 275.6959, | |
| "eval_samples_per_second": 18.611, | |
| "eval_steps_per_second": 2.329, | |
| "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": 1.479651398516736e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |