| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 44.01, |
| "global_step": 1200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0, |
| "learning_rate": 4.9999999999999996e-06, |
| "loss": 11.0188, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.02, |
| "learning_rate": 0.00025, |
| "loss": 8.1559, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.02, |
| "learning_rate": 0.0005, |
| "loss": 6.8029, |
| "step": 100 |
| }, |
| { |
| "epoch": 5.01, |
| "learning_rate": 0.0005833333333333333, |
| "loss": 6.7121, |
| "step": 150 |
| }, |
| { |
| "epoch": 7.01, |
| "learning_rate": 0.0005555555555555556, |
| "loss": 6.6454, |
| "step": 200 |
| }, |
| { |
| "epoch": 9.01, |
| "learning_rate": 0.0005277777777777777, |
| "loss": 6.5652, |
| "step": 250 |
| }, |
| { |
| "epoch": 11.0, |
| "learning_rate": 0.0005, |
| "loss": 6.3824, |
| "step": 300 |
| }, |
| { |
| "epoch": 12.02, |
| "learning_rate": 0.00047222222222222224, |
| "loss": 6.0627, |
| "step": 350 |
| }, |
| { |
| "epoch": 14.02, |
| "learning_rate": 0.00044444444444444436, |
| "loss": 5.9225, |
| "step": 400 |
| }, |
| { |
| "epoch": 16.02, |
| "learning_rate": 0.00041666666666666664, |
| "loss": 5.7523, |
| "step": 450 |
| }, |
| { |
| "epoch": 18.01, |
| "learning_rate": 0.00038888888888888887, |
| "loss": 5.6018, |
| "step": 500 |
| }, |
| { |
| "epoch": 20.01, |
| "learning_rate": 0.0003611111111111111, |
| "loss": 5.4548, |
| "step": 550 |
| }, |
| { |
| "epoch": 22.0, |
| "learning_rate": 0.0003333333333333333, |
| "loss": 5.3217, |
| "step": 600 |
| }, |
| { |
| "epoch": 24.0, |
| "learning_rate": 0.00030555555555555555, |
| "loss": 5.2073, |
| "step": 650 |
| }, |
| { |
| "epoch": 25.02, |
| "learning_rate": 0.0002777777777777778, |
| "loss": 5.0416, |
| "step": 700 |
| }, |
| { |
| "epoch": 27.02, |
| "learning_rate": 0.00025, |
| "loss": 5.0168, |
| "step": 750 |
| }, |
| { |
| "epoch": 29.01, |
| "learning_rate": 0.00022222222222222218, |
| "loss": 4.9351, |
| "step": 800 |
| }, |
| { |
| "epoch": 31.01, |
| "learning_rate": 0.00019444444444444443, |
| "loss": 4.8656, |
| "step": 850 |
| }, |
| { |
| "epoch": 33.01, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 4.8022, |
| "step": 900 |
| }, |
| { |
| "epoch": 35.0, |
| "learning_rate": 0.0001388888888888889, |
| "loss": 4.7449, |
| "step": 950 |
| }, |
| { |
| "epoch": 37.0, |
| "learning_rate": 0.00011111111111111109, |
| "loss": 4.693, |
| "step": 1000 |
| }, |
| { |
| "epoch": 37.0, |
| "eval_loss": 5.544437885284424, |
| "eval_runtime": 2.7184, |
| "eval_samples_per_second": 36.786, |
| "eval_steps_per_second": 2.575, |
| "step": 1000 |
| }, |
| { |
| "epoch": 37.0, |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.544437885284424, |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_ppl": 255.81074283810784, |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_runtime": 2.7184, |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 36.786, |
| "step": 1000 |
| }, |
| { |
| "epoch": 38.02, |
| "learning_rate": 8.333333333333333e-05, |
| "loss": 4.5903, |
| "step": 1050 |
| }, |
| { |
| "epoch": 40.02, |
| "learning_rate": 5.5555555555555545e-05, |
| "loss": 4.6115, |
| "step": 1100 |
| }, |
| { |
| "epoch": 42.01, |
| "learning_rate": 2.7777777777777772e-05, |
| "loss": 4.5829, |
| "step": 1150 |
| }, |
| { |
| "epoch": 44.01, |
| "learning_rate": 0.0, |
| "loss": 4.5627, |
| "step": 1200 |
| } |
| ], |
| "max_steps": 1200, |
| "num_train_epochs": 9223372036854775807, |
| "total_flos": 3.2843363254272e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
