| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 45.00666666666667, |
| "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": 2e-06, |
| "loss": 10.9256, |
| "step": 1 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 9.4777, |
| "step": 50 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 7.5753, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.0, |
| "learning_rate": 0.0003, |
| "loss": 7.4874, |
| "step": 150 |
| }, |
| { |
| "epoch": 8.0, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 7.4256, |
| "step": 200 |
| }, |
| { |
| "epoch": 10.0, |
| "learning_rate": 0.0005, |
| "loss": 7.291, |
| "step": 250 |
| }, |
| { |
| "epoch": 12.0, |
| "learning_rate": 0.0006, |
| "loss": 7.066, |
| "step": 300 |
| }, |
| { |
| "epoch": 14.0, |
| "learning_rate": 0.0005888888888888889, |
| "loss": 6.8676, |
| "step": 350 |
| }, |
| { |
| "epoch": 16.01, |
| "learning_rate": 0.0005777777777777776, |
| "loss": 6.7074, |
| "step": 400 |
| }, |
| { |
| "epoch": 18.01, |
| "learning_rate": 0.0005666666666666666, |
| "loss": 6.577, |
| "step": 450 |
| }, |
| { |
| "epoch": 20.01, |
| "learning_rate": 0.0005555555555555556, |
| "loss": 6.4487, |
| "step": 500 |
| }, |
| { |
| "epoch": 22.01, |
| "learning_rate": 0.0005444444444444444, |
| "loss": 6.3479, |
| "step": 550 |
| }, |
| { |
| "epoch": 24.01, |
| "learning_rate": 0.0005333333333333333, |
| "loss": 6.2602, |
| "step": 600 |
| }, |
| { |
| "epoch": 27.0, |
| "learning_rate": 0.0005222222222222222, |
| "loss": 6.2945, |
| "step": 650 |
| }, |
| { |
| "epoch": 29.0, |
| "learning_rate": 0.0005111111111111111, |
| "loss": 6.1097, |
| "step": 700 |
| }, |
| { |
| "epoch": 31.0, |
| "learning_rate": 0.0005, |
| "loss": 6.049, |
| "step": 750 |
| }, |
| { |
| "epoch": 33.0, |
| "learning_rate": 0.0004888888888888889, |
| "loss": 5.995, |
| "step": 800 |
| }, |
| { |
| "epoch": 35.0, |
| "learning_rate": 0.0004777777777777777, |
| "loss": 5.9402, |
| "step": 850 |
| }, |
| { |
| "epoch": 37.0, |
| "learning_rate": 0.0004666666666666666, |
| "loss": 5.8941, |
| "step": 900 |
| }, |
| { |
| "epoch": 39.0, |
| "learning_rate": 0.00045555555555555556, |
| "loss": 5.8474, |
| "step": 950 |
| }, |
| { |
| "epoch": 41.01, |
| "learning_rate": 0.00044444444444444436, |
| "loss": 5.7977, |
| "step": 1000 |
| }, |
| { |
| "epoch": 41.01, |
| "eval_loss": 6.220484256744385, |
| "eval_runtime": 99.585, |
| "eval_samples_per_second": 129.648, |
| "eval_steps_per_second": 8.104, |
| "step": 1000 |
| }, |
| { |
| "epoch": 41.01, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 6.220484256744385, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 502.9467284033933, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 99.585, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 129.648, |
| "step": 1000 |
| }, |
| { |
| "epoch": 43.01, |
| "learning_rate": 0.0004333333333333333, |
| "loss": 5.745, |
| "step": 1050 |
| }, |
| { |
| "epoch": 45.01, |
| "learning_rate": 0.0004222222222222222, |
| "loss": 5.7002, |
| "step": 1100 |
| } |
| ], |
| "max_steps": 3000, |
| "num_train_epochs": 9223372036854775807, |
| "total_flos": 5.08764100755456e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
