| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 13.038, |
| "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": 11.0014, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.02, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 8.674, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.03, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 6.649, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.05, |
| "learning_rate": 0.0003, |
| "loss": 6.502, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.07, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 6.4531, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.01, |
| "learning_rate": 0.0005, |
| "loss": 6.409, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.03, |
| "learning_rate": 0.0006, |
| "loss": 6.2573, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.04, |
| "learning_rate": 0.0005888888888888889, |
| "loss": 6.0832, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 0.0005777777777777776, |
| "loss": 5.9554, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.0005666666666666666, |
| "loss": 5.873, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.02, |
| "learning_rate": 0.0005555555555555556, |
| "loss": 5.7997, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.04, |
| "learning_rate": 0.0005444444444444444, |
| "loss": 5.7378, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.05, |
| "learning_rate": 0.0005333333333333333, |
| "loss": 5.6778, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.07, |
| "learning_rate": 0.0005222222222222222, |
| "loss": 5.6262, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.01, |
| "learning_rate": 0.0005111111111111111, |
| "loss": 5.5936, |
| "step": 700 |
| }, |
| { |
| "epoch": 3.03, |
| "learning_rate": 0.0005, |
| "loss": 5.5612, |
| "step": 750 |
| }, |
| { |
| "epoch": 3.04, |
| "learning_rate": 0.0004888888888888889, |
| "loss": 5.5263, |
| "step": 800 |
| }, |
| { |
| "epoch": 3.06, |
| "learning_rate": 0.0004777777777777777, |
| "loss": 5.5003, |
| "step": 850 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.0004666666666666666, |
| "loss": 5.4876, |
| "step": 900 |
| }, |
| { |
| "epoch": 4.02, |
| "learning_rate": 0.00045555555555555556, |
| "loss": 5.4625, |
| "step": 950 |
| }, |
| { |
| "epoch": 4.04, |
| "learning_rate": 0.00044444444444444436, |
| "loss": 5.4424, |
| "step": 1000 |
| }, |
| { |
| "epoch": 4.04, |
| "eval_loss": 5.8761162757873535, |
| "eval_runtime": 97.1968, |
| "eval_samples_per_second": 128.749, |
| "eval_steps_per_second": 8.056, |
| "step": 1000 |
| }, |
| { |
| "epoch": 4.04, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.8761162757873535, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 356.4223042778609, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 97.1968, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 128.749, |
| "step": 1000 |
| }, |
| { |
| "epoch": 4.05, |
| "learning_rate": 0.0004333333333333333, |
| "loss": 5.4265, |
| "step": 1050 |
| }, |
| { |
| "epoch": 4.07, |
| "learning_rate": 0.0004222222222222222, |
| "loss": 5.4122, |
| "step": 1100 |
| }, |
| { |
| "epoch": 5.01, |
| "learning_rate": 0.0004111111111111111, |
| "loss": 5.403, |
| "step": 1150 |
| }, |
| { |
| "epoch": 5.03, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 5.3859, |
| "step": 1200 |
| }, |
| { |
| "epoch": 5.05, |
| "learning_rate": 0.00038888888888888887, |
| "loss": 5.3684, |
| "step": 1250 |
| }, |
| { |
| "epoch": 5.06, |
| "learning_rate": 0.00037777777777777777, |
| "loss": 5.3584, |
| "step": 1300 |
| }, |
| { |
| "epoch": 6.01, |
| "learning_rate": 0.00036666666666666667, |
| "loss": 5.3554, |
| "step": 1350 |
| }, |
| { |
| "epoch": 6.02, |
| "learning_rate": 0.0003555555555555555, |
| "loss": 5.341, |
| "step": 1400 |
| }, |
| { |
| "epoch": 6.04, |
| "learning_rate": 0.0003444444444444444, |
| "loss": 5.329, |
| "step": 1450 |
| }, |
| { |
| "epoch": 6.06, |
| "learning_rate": 0.0003333333333333333, |
| "loss": 5.32, |
| "step": 1500 |
| }, |
| { |
| "epoch": 6.07, |
| "learning_rate": 0.0003222222222222222, |
| "loss": 5.3148, |
| "step": 1550 |
| }, |
| { |
| "epoch": 7.02, |
| "learning_rate": 0.00031111111111111107, |
| "loss": 5.3103, |
| "step": 1600 |
| }, |
| { |
| "epoch": 7.03, |
| "learning_rate": 0.0003, |
| "loss": 5.3004, |
| "step": 1650 |
| }, |
| { |
| "epoch": 7.05, |
| "learning_rate": 0.0002888888888888888, |
| "loss": 5.2899, |
| "step": 1700 |
| }, |
| { |
| "epoch": 7.07, |
| "learning_rate": 0.0002777777777777778, |
| "loss": 5.2867, |
| "step": 1750 |
| }, |
| { |
| "epoch": 8.01, |
| "learning_rate": 0.0002666666666666666, |
| "loss": 5.285, |
| "step": 1800 |
| }, |
| { |
| "epoch": 8.02, |
| "learning_rate": 0.00025555555555555553, |
| "loss": 5.2756, |
| "step": 1850 |
| }, |
| { |
| "epoch": 8.04, |
| "learning_rate": 0.00024444444444444443, |
| "loss": 5.2669, |
| "step": 1900 |
| }, |
| { |
| "epoch": 8.06, |
| "learning_rate": 0.0002333333333333333, |
| "loss": 5.2612, |
| "step": 1950 |
| }, |
| { |
| "epoch": 9.0, |
| "learning_rate": 0.00022222222222222218, |
| "loss": 5.2646, |
| "step": 2000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 5.767062187194824, |
| "eval_runtime": 97.7525, |
| "eval_samples_per_second": 128.017, |
| "eval_steps_per_second": 8.01, |
| "step": 2000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.767062187194824, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 319.59743467783994, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 97.7525, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 128.017, |
| "step": 2000 |
| }, |
| { |
| "epoch": 9.02, |
| "learning_rate": 0.0002111111111111111, |
| "loss": 5.2533, |
| "step": 2050 |
| }, |
| { |
| "epoch": 9.03, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 5.2486, |
| "step": 2100 |
| }, |
| { |
| "epoch": 9.05, |
| "learning_rate": 0.00018888888888888888, |
| "loss": 5.2425, |
| "step": 2150 |
| }, |
| { |
| "epoch": 9.07, |
| "learning_rate": 0.00017777777777777776, |
| "loss": 5.2411, |
| "step": 2200 |
| }, |
| { |
| "epoch": 10.01, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 5.2402, |
| "step": 2250 |
| }, |
| { |
| "epoch": 10.03, |
| "learning_rate": 0.00015555555555555554, |
| "loss": 5.2329, |
| "step": 2300 |
| }, |
| { |
| "epoch": 10.04, |
| "learning_rate": 0.0001444444444444444, |
| "loss": 5.2263, |
| "step": 2350 |
| }, |
| { |
| "epoch": 10.06, |
| "learning_rate": 0.0001333333333333333, |
| "loss": 5.2225, |
| "step": 2400 |
| }, |
| { |
| "epoch": 11.0, |
| "learning_rate": 0.00012222222222222221, |
| "loss": 5.2282, |
| "step": 2450 |
| }, |
| { |
| "epoch": 11.02, |
| "learning_rate": 0.00011111111111111109, |
| "loss": 5.2181, |
| "step": 2500 |
| }, |
| { |
| "epoch": 11.04, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 5.2136, |
| "step": 2550 |
| }, |
| { |
| "epoch": 11.05, |
| "learning_rate": 8.888888888888888e-05, |
| "loss": 5.2107, |
| "step": 2600 |
| }, |
| { |
| "epoch": 11.07, |
| "learning_rate": 7.777777777777777e-05, |
| "loss": 5.2093, |
| "step": 2650 |
| }, |
| { |
| "epoch": 12.01, |
| "learning_rate": 6.666666666666666e-05, |
| "loss": 5.2107, |
| "step": 2700 |
| }, |
| { |
| "epoch": 12.03, |
| "learning_rate": 5.5555555555555545e-05, |
| "loss": 5.2057, |
| "step": 2750 |
| }, |
| { |
| "epoch": 12.05, |
| "learning_rate": 4.444444444444444e-05, |
| "loss": 5.1986, |
| "step": 2800 |
| }, |
| { |
| "epoch": 12.06, |
| "learning_rate": 3.333333333333333e-05, |
| "loss": 5.1976, |
| "step": 2850 |
| }, |
| { |
| "epoch": 13.0, |
| "learning_rate": 2.222222222222222e-05, |
| "loss": 5.2036, |
| "step": 2900 |
| }, |
| { |
| "epoch": 13.02, |
| "learning_rate": 1.111111111111111e-05, |
| "loss": 5.1963, |
| "step": 2950 |
| }, |
| { |
| "epoch": 13.04, |
| "learning_rate": 0.0, |
| "loss": 5.1919, |
| "step": 3000 |
| }, |
| { |
| "epoch": 13.04, |
| "eval_loss": 5.735489845275879, |
| "eval_runtime": 96.9335, |
| "eval_samples_per_second": 129.099, |
| "eval_steps_per_second": 8.078, |
| "step": 3000 |
| }, |
| { |
| "epoch": 13.04, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 5.735489845275879, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 309.66462136539093, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 96.9335, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 129.099, |
| "step": 3000 |
| } |
| ], |
| "max_steps": 3000, |
| "num_train_epochs": 9223372036854775807, |
| "total_flos": 1.3382077680451584e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
