| { |
| "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": 10.9249, |
| "step": 1 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 9.4777, |
| "step": 50 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 7.5746, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.0, |
| "learning_rate": 0.0003, |
| "loss": 7.4907, |
| "step": 150 |
| }, |
| { |
| "epoch": 8.0, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 7.4463, |
| "step": 200 |
| }, |
| { |
| "epoch": 10.0, |
| "learning_rate": 0.0005, |
| "loss": 7.3226, |
| "step": 250 |
| }, |
| { |
| "epoch": 12.0, |
| "learning_rate": 0.0006, |
| "loss": 7.1144, |
| "step": 300 |
| }, |
| { |
| "epoch": 14.0, |
| "learning_rate": 0.0005888888888888889, |
| "loss": 6.9433, |
| "step": 350 |
| }, |
| { |
| "epoch": 16.01, |
| "learning_rate": 0.0005777777777777776, |
| "loss": 6.7929, |
| "step": 400 |
| }, |
| { |
| "epoch": 18.01, |
| "learning_rate": 0.0005666666666666666, |
| "loss": 6.6669, |
| "step": 450 |
| }, |
| { |
| "epoch": 20.01, |
| "learning_rate": 0.0005555555555555556, |
| "loss": 6.5658, |
| "step": 500 |
| }, |
| { |
| "epoch": 22.01, |
| "learning_rate": 0.0005444444444444444, |
| "loss": 6.483, |
| "step": 550 |
| }, |
| { |
| "epoch": 24.01, |
| "learning_rate": 0.0005333333333333333, |
| "loss": 6.4096, |
| "step": 600 |
| }, |
| { |
| "epoch": 27.0, |
| "learning_rate": 0.0005222222222222222, |
| "loss": 6.4584, |
| "step": 650 |
| }, |
| { |
| "epoch": 29.0, |
| "learning_rate": 0.0005111111111111111, |
| "loss": 6.2813, |
| "step": 700 |
| }, |
| { |
| "epoch": 31.0, |
| "learning_rate": 0.0005, |
| "loss": 6.2244, |
| "step": 750 |
| }, |
| { |
| "epoch": 33.0, |
| "learning_rate": 0.0004888888888888889, |
| "loss": 6.1753, |
| "step": 800 |
| }, |
| { |
| "epoch": 35.0, |
| "learning_rate": 0.0004777777777777777, |
| "loss": 6.1345, |
| "step": 850 |
| }, |
| { |
| "epoch": 37.0, |
| "learning_rate": 0.0004666666666666666, |
| "loss": 6.088, |
| "step": 900 |
| }, |
| { |
| "epoch": 39.0, |
| "learning_rate": 0.00045555555555555556, |
| "loss": 6.0478, |
| "step": 950 |
| }, |
| { |
| "epoch": 41.01, |
| "learning_rate": 0.00044444444444444436, |
| "loss": 6.0083, |
| "step": 1000 |
| }, |
| { |
| "epoch": 41.01, |
| "eval_loss": 6.419702053070068, |
| "eval_runtime": 96.6209, |
| "eval_samples_per_second": 133.625, |
| "eval_steps_per_second": 8.352, |
| "step": 1000 |
| }, |
| { |
| "epoch": 41.01, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 6.419702053070068, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 613.8202010332798, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 96.6209, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 133.625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 43.01, |
| "learning_rate": 0.0004333333333333333, |
| "loss": 5.9754, |
| "step": 1050 |
| }, |
| { |
| "epoch": 45.01, |
| "learning_rate": 0.0004222222222222222, |
| "loss": 5.9343, |
| "step": 1100 |
| }, |
| { |
| "epoch": 47.01, |
| "learning_rate": 0.0004111111111111111, |
| "loss": 5.8889, |
| "step": 1150 |
| }, |
| { |
| "epoch": 49.01, |
| "learning_rate": 0.00039999999999999996, |
| "loss": 5.8478, |
| "step": 1200 |
| }, |
| { |
| "epoch": 52.0, |
| "learning_rate": 0.00038888888888888887, |
| "loss": 5.9313, |
| "step": 1250 |
| }, |
| { |
| "epoch": 54.0, |
| "learning_rate": 0.00037777777777777777, |
| "loss": 5.781, |
| "step": 1300 |
| }, |
| { |
| "epoch": 56.0, |
| "learning_rate": 0.00036666666666666667, |
| "loss": 5.7562, |
| "step": 1350 |
| }, |
| { |
| "epoch": 58.0, |
| "learning_rate": 0.0003555555555555555, |
| "loss": 5.728, |
| "step": 1400 |
| }, |
| { |
| "epoch": 60.0, |
| "learning_rate": 0.0003444444444444444, |
| "loss": 5.7015, |
| "step": 1450 |
| }, |
| { |
| "epoch": 62.0, |
| "learning_rate": 0.0003333333333333333, |
| "loss": 5.6735, |
| "step": 1500 |
| }, |
| { |
| "epoch": 64.0, |
| "learning_rate": 0.0003222222222222222, |
| "loss": 5.6514, |
| "step": 1550 |
| }, |
| { |
| "epoch": 66.01, |
| "learning_rate": 0.00031111111111111107, |
| "loss": 5.6291, |
| "step": 1600 |
| }, |
| { |
| "epoch": 68.01, |
| "learning_rate": 0.0003, |
| "loss": 5.6076, |
| "step": 1650 |
| }, |
| { |
| "epoch": 70.01, |
| "learning_rate": 0.0002888888888888888, |
| "loss": 5.588, |
| "step": 1700 |
| }, |
| { |
| "epoch": 72.01, |
| "learning_rate": 0.0002777777777777778, |
| "loss": 5.5709, |
| "step": 1750 |
| }, |
| { |
| "epoch": 74.01, |
| "learning_rate": 0.0002666666666666666, |
| "loss": 5.5555, |
| "step": 1800 |
| }, |
| { |
| "epoch": 77.0, |
| "learning_rate": 0.00025555555555555553, |
| "loss": 5.6504, |
| "step": 1850 |
| }, |
| { |
| "epoch": 79.0, |
| "learning_rate": 0.00024444444444444443, |
| "loss": 5.5204, |
| "step": 1900 |
| }, |
| { |
| "epoch": 81.0, |
| "learning_rate": 0.0002333333333333333, |
| "loss": 5.503, |
| "step": 1950 |
| }, |
| { |
| "epoch": 83.0, |
| "learning_rate": 0.00022222222222222218, |
| "loss": 5.4859, |
| "step": 2000 |
| }, |
| { |
| "epoch": 83.0, |
| "eval_loss": 6.717416286468506, |
| "eval_runtime": 98.5086, |
| "eval_samples_per_second": 131.065, |
| "eval_steps_per_second": 8.192, |
| "step": 2000 |
| }, |
| { |
| "epoch": 83.0, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 6.717416286468506, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 826.6788484978945, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 98.5086, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 131.065, |
| "step": 2000 |
| }, |
| { |
| "epoch": 85.0, |
| "learning_rate": 0.0002111111111111111, |
| "loss": 5.4736, |
| "step": 2050 |
| }, |
| { |
| "epoch": 87.0, |
| "learning_rate": 0.00019999999999999998, |
| "loss": 5.4581, |
| "step": 2100 |
| }, |
| { |
| "epoch": 89.0, |
| "learning_rate": 0.00018888888888888888, |
| "loss": 5.4442, |
| "step": 2150 |
| }, |
| { |
| "epoch": 91.01, |
| "learning_rate": 0.00017777777777777776, |
| "loss": 5.4336, |
| "step": 2200 |
| }, |
| { |
| "epoch": 93.01, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 5.4204, |
| "step": 2250 |
| }, |
| { |
| "epoch": 95.01, |
| "learning_rate": 0.00015555555555555554, |
| "loss": 5.4075, |
| "step": 2300 |
| }, |
| { |
| "epoch": 97.01, |
| "learning_rate": 0.0001444444444444444, |
| "loss": 5.3961, |
| "step": 2350 |
| }, |
| { |
| "epoch": 99.01, |
| "learning_rate": 0.0001333333333333333, |
| "loss": 5.386, |
| "step": 2400 |
| }, |
| { |
| "epoch": 102.0, |
| "learning_rate": 0.00012222222222222221, |
| "loss": 5.4928, |
| "step": 2450 |
| }, |
| { |
| "epoch": 104.0, |
| "learning_rate": 0.00011111111111111109, |
| "loss": 5.3674, |
| "step": 2500 |
| }, |
| { |
| "epoch": 106.0, |
| "learning_rate": 9.999999999999999e-05, |
| "loss": 5.3581, |
| "step": 2550 |
| }, |
| { |
| "epoch": 108.0, |
| "learning_rate": 8.888888888888888e-05, |
| "loss": 5.3486, |
| "step": 2600 |
| }, |
| { |
| "epoch": 110.0, |
| "learning_rate": 7.777777777777777e-05, |
| "loss": 5.3394, |
| "step": 2650 |
| }, |
| { |
| "epoch": 112.0, |
| "learning_rate": 6.666666666666666e-05, |
| "loss": 5.3318, |
| "step": 2700 |
| }, |
| { |
| "epoch": 114.0, |
| "learning_rate": 5.5555555555555545e-05, |
| "loss": 5.3247, |
| "step": 2750 |
| }, |
| { |
| "epoch": 116.01, |
| "learning_rate": 4.444444444444444e-05, |
| "loss": 5.3189, |
| "step": 2800 |
| }, |
| { |
| "epoch": 118.01, |
| "learning_rate": 3.333333333333333e-05, |
| "loss": 5.3142, |
| "step": 2850 |
| }, |
| { |
| "epoch": 120.01, |
| "learning_rate": 2.222222222222222e-05, |
| "loss": 5.3077, |
| "step": 2900 |
| }, |
| { |
| "epoch": 122.01, |
| "learning_rate": 1.111111111111111e-05, |
| "loss": 5.3031, |
| "step": 2950 |
| }, |
| { |
| "epoch": 124.01, |
| "learning_rate": 0.0, |
| "loss": 5.2998, |
| "step": 3000 |
| }, |
| { |
| "epoch": 124.01, |
| "eval_loss": 6.8880181312561035, |
| "eval_runtime": 98.4154, |
| "eval_samples_per_second": 131.189, |
| "eval_steps_per_second": 8.2, |
| "step": 3000 |
| }, |
| { |
| "epoch": 124.01, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_loss": 6.8880181312561035, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_ppl": 980.456354621549, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_runtime": 98.4154, |
| "eval_/home/s2678328/mission-impossible-language-models/training/babylm_dataset.py_samples_per_second": 131.189, |
| "step": 3000 |
| } |
| ], |
| "max_steps": 3000, |
| "num_train_epochs": 9223372036854775807, |
| "total_flos": 1.3880528524541952e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
