| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.564373897707231, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02821869488536155, | |
| "grad_norm": 2.359462261199951, | |
| "learning_rate": 1e-05, | |
| "loss": 6.2205, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02821869488536155, | |
| "eval_loss": 6.2788615226745605, | |
| "eval_runtime": 36.398, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.824, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0564373897707231, | |
| "grad_norm": 2.0334458351135254, | |
| "learning_rate": 2e-05, | |
| "loss": 6.4757, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08465608465608465, | |
| "grad_norm": 2.242079973220825, | |
| "learning_rate": 3e-05, | |
| "loss": 6.3527, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1128747795414462, | |
| "grad_norm": 2.256972312927246, | |
| "learning_rate": 4e-05, | |
| "loss": 6.5204, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.14109347442680775, | |
| "grad_norm": 2.1701977252960205, | |
| "learning_rate": 5e-05, | |
| "loss": 6.0037, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.14109347442680775, | |
| "eval_loss": 6.136302471160889, | |
| "eval_runtime": 0.4876, | |
| "eval_samples_per_second": 123.04, | |
| "eval_steps_per_second": 61.52, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1693121693121693, | |
| "grad_norm": 1.918559193611145, | |
| "learning_rate": 6e-05, | |
| "loss": 6.5218, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 2.5789589881896973, | |
| "learning_rate": 7e-05, | |
| "loss": 6.2126, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2257495590828924, | |
| "grad_norm": 2.5044870376586914, | |
| "learning_rate": 8e-05, | |
| "loss": 6.2001, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.25396825396825395, | |
| "grad_norm": 2.7968616485595703, | |
| "learning_rate": 9e-05, | |
| "loss": 6.5004, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2821869488536155, | |
| "grad_norm": 2.528766393661499, | |
| "learning_rate": 0.0001, | |
| "loss": 6.0157, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2821869488536155, | |
| "eval_loss": 5.689663887023926, | |
| "eval_runtime": 0.4947, | |
| "eval_samples_per_second": 121.289, | |
| "eval_steps_per_second": 60.645, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31040564373897706, | |
| "grad_norm": 3.115682601928711, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 6.0316, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3386243386243386, | |
| "grad_norm": 3.406926155090332, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 6.203, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.36684303350970016, | |
| "grad_norm": 3.4907941818237305, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 5.9364, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 3.92707896232605, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 5.4231, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.42328042328042326, | |
| "grad_norm": 3.5260536670684814, | |
| "learning_rate": 5e-05, | |
| "loss": 5.4519, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.42328042328042326, | |
| "eval_loss": 4.813125133514404, | |
| "eval_runtime": 0.5202, | |
| "eval_samples_per_second": 115.351, | |
| "eval_steps_per_second": 57.676, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4514991181657848, | |
| "grad_norm": 3.738027811050415, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 4.8855, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.47971781305114636, | |
| "grad_norm": 3.98640775680542, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 4.7694, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5079365079365079, | |
| "grad_norm": 3.7007837295532227, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 4.4581, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5361552028218695, | |
| "grad_norm": 3.676358699798584, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 4.7911, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.564373897707231, | |
| "grad_norm": 3.9247279167175293, | |
| "learning_rate": 0.0, | |
| "loss": 5.0012, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.564373897707231, | |
| "eval_loss": 4.537546634674072, | |
| "eval_runtime": 0.488, | |
| "eval_samples_per_second": 122.95, | |
| "eval_steps_per_second": 61.475, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2046335826001920.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |