| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 25.020833333333332, | |
| "global_step": 1100, | |
| "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": 10.9692, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00025, | |
| "loss": 8.063, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0005, | |
| "loss": 4.9432, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.0005833333333333333, | |
| "loss": 3.9936, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.0005555555555555556, | |
| "loss": 3.7529, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.0005277777777777777, | |
| "loss": 3.6008, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.0005, | |
| "loss": 3.3332, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 0.00047222222222222224, | |
| "loss": 3.089, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 0.00044444444444444436, | |
| "loss": 2.8687, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 0.00041666666666666664, | |
| "loss": 2.7458, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 0.00038888888888888887, | |
| "loss": 2.6512, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 0.0003611111111111111, | |
| "loss": 2.5787, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 2.5183, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 0.00030555555555555555, | |
| "loss": 2.4963, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 2.4139, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 0.00025, | |
| "loss": 2.3755, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 0.00022222222222222218, | |
| "loss": 2.3405, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 0.00019444444444444443, | |
| "loss": 2.3063, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 20.03, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 2.2768, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 2.2799, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 0.00011111111111111109, | |
| "loss": 2.2247, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "eval_loss": 2.6665191650390625, | |
| "eval_runtime": 3.9154, | |
| "eval_samples_per_second": 41.631, | |
| "eval_steps_per_second": 2.809, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_loss": 2.6665191650390625, | |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_ppl": 14.389793420654337, | |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_runtime": 3.9154, | |
| "eval_/local/xiulyang/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 41.631, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 2.202, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "learning_rate": 5.5555555555555545e-05, | |
| "loss": 2.1818, | |
| "step": 1100 | |
| } | |
| ], | |
| "max_steps": 1200, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 2.988527616e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |