| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 124.008, |
| "global_step": 3000, |
| "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": 10.9377, |
| "step": 1 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 9.4009, |
| "step": 50 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 7.1915, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.0, |
| "learning_rate": 0.0003, |
| "loss": 6.4074, |
| "step": 150 |
| }, |
| { |
| "epoch": 8.0, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 5.6889, |
| "step": 200 |
| }, |
| { |
| "epoch": 10.0, |
| "learning_rate": 0.0005, |
| "loss": 5.326, |
| "step": 250 |
| }, |
| { |
| "epoch": 12.0, |
| "learning_rate": 0.0006, |
| "loss": 5.0873, |
| "step": 300 |
| }, |
| { |
| "epoch": 14.0, |
| "learning_rate": 0.0005888888888888889, |
| "loss": 4.7619, |
| "step": 350 |
| }, |
| { |
| "epoch": 16.01, |
| "learning_rate": 0.0005777777777777776, |
| "loss": 4.4374, |
| "step": 400 |
| }, |
| { |
| "epoch": 18.01, |
| "learning_rate": 0.0005666666666666666, |
| "loss": 4.2331, |
| "step": 450 |
| }, |
| { |
| "epoch": 20.01, |
| "learning_rate": 0.0005555555555555556, |
| "loss": 4.0853, |
| "step": 500 |
| }, |
| { |
| "epoch": 22.01, |
| "learning_rate": 0.0005444444444444444, |
| "loss": 3.9849, |
| "step": 550 |
| }, |
| { |
| "epoch": 24.01, |
| "learning_rate": 0.0005333333333333333, |
| "loss": 3.8903, |
| "step": 600 |
| }, |
| { |
| "epoch": 27.0, |
| "learning_rate": 0.0005222222222222222, |
| "loss": 3.8802, |
| "step": 650 |
| }, |
| { |
| "epoch": 29.0, |
| "learning_rate": 0.0005111111111111111, |
| "loss": 3.748, |
| "step": 700 |
| }, |
| { |
| "epoch": 31.0, |
| "learning_rate": 0.0005, |
| "loss": 3.6859, |
| "step": 750 |
| }, |
| { |
| "epoch": 33.0, |
| "learning_rate": 0.0004888888888888889, |
| "loss": 3.6323, |
| "step": 800 |
| }, |
| { |
| "epoch": 35.0, |
| "learning_rate": 0.0004777777777777777, |
| "loss": 3.5863, |
| "step": 850 |
| }, |
| { |
| "epoch": 37.0, |
| "learning_rate": 0.0004666666666666666, |
| "loss": 3.5434, |
| "step": 900 |
| }, |
| { |
| "epoch": 39.0, |
| "learning_rate": 0.00045555555555555556, |
| "loss": 3.5047, |
| "step": 950 |
| }, |
| { |
| "epoch": 41.01, |
| "learning_rate": 0.00044444444444444436, |
| "loss": 3.4702, |
| "step": 1000 |
| }, |
| { |
| "epoch": 41.01, |
| "eval_loss": 5.252923965454102, |
| "eval_runtime": 95.4872, |
| "eval_samples_per_second": 135.212, |
| "eval_steps_per_second": 8.451, |
| "step": 1000 |
| }, |
| { |
| "epoch": 41.01, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.252923965454102, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 191.12429306908916, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 95.4872, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 135.212, |
| "step": 1000 |
| }, |
| { |
| "epoch": 43.01, |
| "learning_rate": 0.0004333333333333333, |
| "loss": 3.4386, |
| "step": 1050 |
| }, |
| { |
| "epoch": 45.01, |
| "learning_rate": 0.0004222222222222222, |
| "loss": 3.4071, |
| "step": 1100 |
| }, |
| { |
| "epoch": 47.01, |
| "learning_rate": 0.0004111111111111111, |
| "loss": 3.3787, |
| "step": 1150 |
| }, |
| { |
| "epoch": 49.01, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 3.354, |
| "step": 1200 |
| }, |
| { |
| "epoch": 52.0, |
| "learning_rate": 0.00038888888888888887, |
| "loss": 3.3887, |
| "step": 1250 |
| }, |
| { |
| "epoch": 54.0, |
| "learning_rate": 0.00037777777777777777, |
| "loss": 3.3078, |
| "step": 1300 |
| }, |
| { |
| "epoch": 56.0, |
| "learning_rate": 0.00036666666666666667, |
| "loss": 3.286, |
| "step": 1350 |
| }, |
| { |
| "epoch": 58.0, |
| "learning_rate": 0.0003555555555555555, |
| "loss": 3.2661, |
| "step": 1400 |
| }, |
| { |
| "epoch": 60.0, |
| "learning_rate": 0.0003444444444444444, |
| "loss": 3.2487, |
| "step": 1450 |
| }, |
| { |
| "epoch": 62.0, |
| "learning_rate": 0.0003333333333333333, |
| "loss": 3.2283, |
| "step": 1500 |
| }, |
| { |
| "epoch": 64.0, |
| "learning_rate": 0.0003222222222222222, |
| "loss": 3.2112, |
| "step": 1550 |
| }, |
| { |
| "epoch": 66.01, |
| "learning_rate": 0.00031111111111111107, |
| "loss": 3.1923, |
| "step": 1600 |
| }, |
| { |
| "epoch": 68.01, |
| "learning_rate": 0.0003, |
| "loss": 3.1748, |
| "step": 1650 |
| }, |
| { |
| "epoch": 70.01, |
| "learning_rate": 0.0002888888888888888, |
| "loss": 3.1582, |
| "step": 1700 |
| }, |
| { |
| "epoch": 72.01, |
| "learning_rate": 0.0002777777777777778, |
| "loss": 3.1442, |
| "step": 1750 |
| }, |
| { |
| "epoch": 74.01, |
| "learning_rate": 0.0002666666666666666, |
| "loss": 3.1304, |
| "step": 1800 |
| }, |
| { |
| "epoch": 77.0, |
| "learning_rate": 0.00025555555555555553, |
| "loss": 3.1727, |
| "step": 1850 |
| }, |
| { |
| "epoch": 79.0, |
| "learning_rate": 0.00024444444444444443, |
| "loss": 3.1049, |
| "step": 1900 |
| }, |
| { |
| "epoch": 81.0, |
| "learning_rate": 0.0002333333333333333, |
| "loss": 3.0929, |
| "step": 1950 |
| }, |
| { |
| "epoch": 83.0, |
| "learning_rate": 0.00022222222222222218, |
| "loss": 3.0823, |
| "step": 2000 |
| }, |
| { |
| "epoch": 83.0, |
| "eval_loss": 5.627013683319092, |
| "eval_runtime": 97.1891, |
| "eval_samples_per_second": 132.844, |
| "eval_steps_per_second": 8.303, |
| "step": 2000 |
| }, |
| { |
| "epoch": 83.0, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.627013683319092, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 277.83118563338775, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 97.1891, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 132.844, |
| "step": 2000 |
| }, |
| { |
| "epoch": 85.0, |
| "learning_rate": 0.0002111111111111111, |
| "loss": 3.0729, |
| "step": 2050 |
| }, |
| { |
| "epoch": 87.0, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 3.0621, |
| "step": 2100 |
| }, |
| { |
| "epoch": 89.0, |
| "learning_rate": 0.00018888888888888888, |
| "loss": 3.0523, |
| "step": 2150 |
| }, |
| { |
| "epoch": 91.01, |
| "learning_rate": 0.00017777777777777776, |
| "loss": 3.0421, |
| "step": 2200 |
| }, |
| { |
| "epoch": 93.01, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 3.0323, |
| "step": 2250 |
| }, |
| { |
| "epoch": 95.01, |
| "learning_rate": 0.00015555555555555554, |
| "loss": 3.0235, |
| "step": 2300 |
| }, |
| { |
| "epoch": 97.01, |
| "learning_rate": 0.0001444444444444444, |
| "loss": 3.0148, |
| "step": 2350 |
| }, |
| { |
| "epoch": 99.01, |
| "learning_rate": 0.0001333333333333333, |
| "loss": 3.0069, |
| "step": 2400 |
| }, |
| { |
| "epoch": 102.0, |
| "learning_rate": 0.00012222222222222221, |
| "loss": 3.0535, |
| "step": 2450 |
| }, |
| { |
| "epoch": 104.0, |
| "learning_rate": 0.00011111111111111109, |
| "loss": 2.9932, |
| "step": 2500 |
| }, |
| { |
| "epoch": 106.0, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 2.9858, |
| "step": 2550 |
| }, |
| { |
| "epoch": 108.0, |
| "learning_rate": 8.888888888888888e-05, |
| "loss": 2.9793, |
| "step": 2600 |
| }, |
| { |
| "epoch": 110.0, |
| "learning_rate": 7.777777777777777e-05, |
| "loss": 2.9737, |
| "step": 2650 |
| }, |
| { |
| "epoch": 112.0, |
| "learning_rate": 6.666666666666666e-05, |
| "loss": 2.9676, |
| "step": 2700 |
| }, |
| { |
| "epoch": 114.0, |
| "learning_rate": 5.5555555555555545e-05, |
| "loss": 2.9624, |
| "step": 2750 |
| }, |
| { |
| "epoch": 116.01, |
| "learning_rate": 4.444444444444444e-05, |
| "loss": 2.9581, |
| "step": 2800 |
| }, |
| { |
| "epoch": 118.01, |
| "learning_rate": 3.333333333333333e-05, |
| "loss": 2.9534, |
| "step": 2850 |
| }, |
| { |
| "epoch": 120.01, |
| "learning_rate": 2.222222222222222e-05, |
| "loss": 2.9494, |
| "step": 2900 |
| }, |
| { |
| "epoch": 122.01, |
| "learning_rate": 1.111111111111111e-05, |
| "loss": 2.946, |
| "step": 2950 |
| }, |
| { |
| "epoch": 124.01, |
| "learning_rate": 0.0, |
| "loss": 2.9434, |
| "step": 3000 |
| }, |
| { |
| "epoch": 124.01, |
| "eval_loss": 5.734775543212891, |
| "eval_runtime": 97.6358, |
| "eval_samples_per_second": 132.236, |
| "eval_steps_per_second": 8.265, |
| "step": 3000 |
| }, |
| { |
| "epoch": 124.01, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.734775543212891, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 309.44350626840173, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 97.6358, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 132.236, |
| "step": 3000 |
| } |
| ], |
| "max_steps": 3000, |
| "num_train_epochs": 9223372036854775807, |
| "total_flos": 1.3880528524541952e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
