| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.190476190476191, | |
| "eval_steps": 500, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 9600.0, | |
| "learning_rate": 0.0, | |
| "loss": 55.1354, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 9600.0, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 55.1354, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 16384.0, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 54.5963, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 908.0, | |
| "learning_rate": 2e-05, | |
| "loss": 27.4621, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 792.0, | |
| "learning_rate": 1.989821441880933e-05, | |
| "loss": 24.9079, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 468.0, | |
| "learning_rate": 1.9594929736144978e-05, | |
| "loss": 8.0325, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 486.0, | |
| "learning_rate": 1.9096319953545186e-05, | |
| "loss": 22.7508, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.380952380952381, | |
| "grad_norm": 1256.0, | |
| "learning_rate": 1.8412535328311813e-05, | |
| "loss": 23.4299, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 302.0, | |
| "learning_rate": 1.7557495743542586e-05, | |
| "loss": 19.6502, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.7619047619047619, | |
| "grad_norm": 410.0, | |
| "learning_rate": 1.6548607339452853e-05, | |
| "loss": 18.8854, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9523809523809523, | |
| "grad_norm": 470.0, | |
| "learning_rate": 1.5406408174555978e-05, | |
| "loss": 18.1702, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 62.25, | |
| "learning_rate": 1.4154150130018867e-05, | |
| "loss": 4.2635, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 888.0, | |
| "learning_rate": 1.2817325568414299e-05, | |
| "loss": 17.1308, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 492.0, | |
| "learning_rate": 1.1423148382732854e-05, | |
| "loss": 15.8069, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 712.0, | |
| "learning_rate": 1e-05, | |
| "loss": 15.6847, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 760.0, | |
| "learning_rate": 8.576851617267151e-06, | |
| "loss": 15.4479, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 528.0, | |
| "learning_rate": 7.182674431585703e-06, | |
| "loss": 14.7789, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 76.0, | |
| "learning_rate": 5.845849869981137e-06, | |
| "loss": 3.5908, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.1904761904761907, | |
| "grad_norm": 458.0, | |
| "learning_rate": 4.593591825444028e-06, | |
| "loss": 14.2981, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.380952380952381, | |
| "grad_norm": 480.0, | |
| "learning_rate": 3.4513926605471504e-06, | |
| "loss": 14.2165, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 476.0, | |
| "learning_rate": 2.4425042564574186e-06, | |
| "loss": 14.1534, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.761904761904762, | |
| "grad_norm": 426.0, | |
| "learning_rate": 1.587464671688187e-06, | |
| "loss": 14.0626, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.9523809523809526, | |
| "grad_norm": 418.0, | |
| "learning_rate": 9.036800464548157e-07, | |
| "loss": 14.0506, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 99.0, | |
| "learning_rate": 4.0507026385502747e-07, | |
| "loss": 3.5034, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.190476190476191, | |
| "grad_norm": 384.0, | |
| "learning_rate": 1.0178558119067316e-07, | |
| "loss": 13.9924, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 377332860715008.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |