| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2846975088967972, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014234875444839857, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 3346.0093, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.014234875444839857, | |
| "eval_loss": NaN, | |
| "eval_runtime": 69.0908, | |
| "eval_samples_per_second": 6.861, | |
| "eval_steps_per_second": 0.868, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.028469750889679714, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 5441.3306, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.042704626334519574, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 850.3894, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05693950177935943, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 1138.0363, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0711743772241993, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 3671.7151, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0711743772241993, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.5954, | |
| "eval_samples_per_second": 37.633, | |
| "eval_steps_per_second": 4.764, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08540925266903915, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 2033.0361, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.099644128113879, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 366.2248, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.11387900355871886, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 588.2904, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.12811387900355872, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 2884.811, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1423487544483986, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 441.9196, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1423487544483986, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.6367, | |
| "eval_samples_per_second": 37.51, | |
| "eval_steps_per_second": 4.748, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15658362989323843, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2346.3477, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1708185053380783, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 362.8186, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.18505338078291814, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2926.6011, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.199288256227758, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 595.2302, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.21352313167259787, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 591.5734, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.21352313167259787, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.5898, | |
| "eval_samples_per_second": 37.649, | |
| "eval_steps_per_second": 4.766, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2277580071174377, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1069.6057, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.24199288256227758, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1501.6047, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.25622775800711745, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 737.9307, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2704626334519573, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 223.4566, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2846975088967972, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 1692.3673, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2846975088967972, | |
| "eval_loss": NaN, | |
| "eval_runtime": 12.5841, | |
| "eval_samples_per_second": 37.667, | |
| "eval_steps_per_second": 4.768, | |
| "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": 1.0247529975250944e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |