|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 45.625, |
|
"eval_steps": 500, |
|
"global_step": 730, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.625, |
|
"grad_norm": 8.0, |
|
"learning_rate": 1e-05, |
|
"loss": 12.2241, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"grad_norm": 6.125, |
|
"learning_rate": 2e-05, |
|
"loss": 11.9861, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 1.875, |
|
"grad_norm": 8.0625, |
|
"learning_rate": 3e-05, |
|
"loss": 11.267, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"grad_norm": 7.1875, |
|
"learning_rate": 4e-05, |
|
"loss": 10.0343, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 3.125, |
|
"grad_norm": 6.8125, |
|
"learning_rate": 5e-05, |
|
"loss": 8.5167, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 3.75, |
|
"grad_norm": 7.90625, |
|
"learning_rate": 6e-05, |
|
"loss": 7.3404, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 4.375, |
|
"grad_norm": 10.0625, |
|
"learning_rate": 7e-05, |
|
"loss": 6.344, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"grad_norm": 12.875, |
|
"learning_rate": 8e-05, |
|
"loss": 5.2821, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 5.625, |
|
"grad_norm": 14.375, |
|
"learning_rate": 9e-05, |
|
"loss": 3.6695, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 6.25, |
|
"grad_norm": 14.6875, |
|
"learning_rate": 0.0001, |
|
"loss": 2.1107, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 6.875, |
|
"grad_norm": 2.5, |
|
"learning_rate": 9.857142857142858e-05, |
|
"loss": 0.6406, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 7.5, |
|
"grad_norm": 1.453125, |
|
"learning_rate": 9.714285714285715e-05, |
|
"loss": 0.3052, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 8.125, |
|
"grad_norm": 1.4765625, |
|
"learning_rate": 9.571428571428573e-05, |
|
"loss": 0.3007, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 8.75, |
|
"grad_norm": 1.8359375, |
|
"learning_rate": 9.428571428571429e-05, |
|
"loss": 0.2633, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 9.375, |
|
"grad_norm": 25.75, |
|
"learning_rate": 9.285714285714286e-05, |
|
"loss": 0.2614, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"grad_norm": 0.89453125, |
|
"learning_rate": 9.142857142857143e-05, |
|
"loss": 0.2326, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 10.625, |
|
"grad_norm": 1.9453125, |
|
"learning_rate": 9e-05, |
|
"loss": 0.2548, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 11.25, |
|
"grad_norm": 1.6171875, |
|
"learning_rate": 8.857142857142857e-05, |
|
"loss": 0.232, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 11.875, |
|
"grad_norm": 0.8515625, |
|
"learning_rate": 8.714285714285715e-05, |
|
"loss": 0.2321, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 12.5, |
|
"grad_norm": 1.171875, |
|
"learning_rate": 8.571428571428571e-05, |
|
"loss": 0.2136, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 13.125, |
|
"grad_norm": 1.1015625, |
|
"learning_rate": 8.428571428571429e-05, |
|
"loss": 0.2142, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 13.75, |
|
"grad_norm": 1.7109375, |
|
"learning_rate": 8.285714285714287e-05, |
|
"loss": 0.2018, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 14.375, |
|
"grad_norm": 0.76171875, |
|
"learning_rate": 8.142857142857143e-05, |
|
"loss": 0.1877, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"grad_norm": 0.875, |
|
"learning_rate": 8e-05, |
|
"loss": 0.1963, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 15.625, |
|
"grad_norm": 1.71875, |
|
"learning_rate": 7.857142857142858e-05, |
|
"loss": 0.1883, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 16.25, |
|
"grad_norm": 0.8671875, |
|
"learning_rate": 7.714285714285715e-05, |
|
"loss": 0.1972, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 16.875, |
|
"grad_norm": 2.953125, |
|
"learning_rate": 7.571428571428571e-05, |
|
"loss": 0.1741, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 17.5, |
|
"grad_norm": 0.94921875, |
|
"learning_rate": 7.428571428571429e-05, |
|
"loss": 0.1656, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 18.125, |
|
"grad_norm": 0.734375, |
|
"learning_rate": 7.285714285714286e-05, |
|
"loss": 0.1677, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 18.75, |
|
"grad_norm": 0.91796875, |
|
"learning_rate": 7.142857142857143e-05, |
|
"loss": 0.1572, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 19.375, |
|
"grad_norm": 1.359375, |
|
"learning_rate": 7e-05, |
|
"loss": 0.1567, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"grad_norm": 1.4375, |
|
"learning_rate": 6.857142857142858e-05, |
|
"loss": 0.1522, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 20.625, |
|
"grad_norm": 0.55859375, |
|
"learning_rate": 6.714285714285714e-05, |
|
"loss": 0.144, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 21.25, |
|
"grad_norm": 0.51953125, |
|
"learning_rate": 6.571428571428571e-05, |
|
"loss": 0.1488, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 21.875, |
|
"grad_norm": 0.89453125, |
|
"learning_rate": 6.428571428571429e-05, |
|
"loss": 0.1504, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 22.5, |
|
"grad_norm": 0.5078125, |
|
"learning_rate": 6.285714285714286e-05, |
|
"loss": 0.1332, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 23.125, |
|
"grad_norm": 0.6796875, |
|
"learning_rate": 6.142857142857143e-05, |
|
"loss": 0.136, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 23.75, |
|
"grad_norm": 0.66015625, |
|
"learning_rate": 6e-05, |
|
"loss": 0.135, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 24.375, |
|
"grad_norm": 0.78515625, |
|
"learning_rate": 5.8571428571428575e-05, |
|
"loss": 0.1264, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"grad_norm": 0.5703125, |
|
"learning_rate": 5.714285714285714e-05, |
|
"loss": 0.1298, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 25.625, |
|
"grad_norm": 0.5546875, |
|
"learning_rate": 5.571428571428572e-05, |
|
"loss": 0.1219, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 26.25, |
|
"grad_norm": 0.65625, |
|
"learning_rate": 5.428571428571428e-05, |
|
"loss": 0.1254, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 26.875, |
|
"grad_norm": 0.73828125, |
|
"learning_rate": 5.285714285714286e-05, |
|
"loss": 0.1229, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 27.5, |
|
"grad_norm": 0.6484375, |
|
"learning_rate": 5.142857142857143e-05, |
|
"loss": 0.1166, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 28.125, |
|
"grad_norm": 0.66796875, |
|
"learning_rate": 5e-05, |
|
"loss": 0.1178, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 28.75, |
|
"grad_norm": 0.82421875, |
|
"learning_rate": 4.8571428571428576e-05, |
|
"loss": 0.1141, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 29.375, |
|
"grad_norm": 0.62109375, |
|
"learning_rate": 4.714285714285714e-05, |
|
"loss": 0.1138, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"grad_norm": 0.765625, |
|
"learning_rate": 4.5714285714285716e-05, |
|
"loss": 0.1149, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 30.625, |
|
"grad_norm": 0.546875, |
|
"learning_rate": 4.428571428571428e-05, |
|
"loss": 0.1043, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 31.25, |
|
"grad_norm": 0.515625, |
|
"learning_rate": 4.2857142857142856e-05, |
|
"loss": 0.1075, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 31.875, |
|
"grad_norm": 0.66796875, |
|
"learning_rate": 4.1428571428571437e-05, |
|
"loss": 0.107, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 32.5, |
|
"grad_norm": 0.75, |
|
"learning_rate": 4e-05, |
|
"loss": 0.104, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 33.125, |
|
"grad_norm": 0.5859375, |
|
"learning_rate": 3.857142857142858e-05, |
|
"loss": 0.1089, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 33.75, |
|
"grad_norm": 0.69140625, |
|
"learning_rate": 3.7142857142857143e-05, |
|
"loss": 0.1001, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 34.375, |
|
"grad_norm": 0.58203125, |
|
"learning_rate": 3.571428571428572e-05, |
|
"loss": 0.1043, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 35.0, |
|
"grad_norm": 0.66015625, |
|
"learning_rate": 3.428571428571429e-05, |
|
"loss": 0.101, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 35.625, |
|
"grad_norm": 0.6953125, |
|
"learning_rate": 3.285714285714286e-05, |
|
"loss": 0.0987, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 36.25, |
|
"grad_norm": 0.62890625, |
|
"learning_rate": 3.142857142857143e-05, |
|
"loss": 0.0976, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 36.875, |
|
"grad_norm": 0.82421875, |
|
"learning_rate": 3e-05, |
|
"loss": 0.0984, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 37.5, |
|
"grad_norm": 0.703125, |
|
"learning_rate": 2.857142857142857e-05, |
|
"loss": 0.0891, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 38.125, |
|
"grad_norm": 0.60546875, |
|
"learning_rate": 2.714285714285714e-05, |
|
"loss": 0.0943, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 38.75, |
|
"grad_norm": 0.63671875, |
|
"learning_rate": 2.5714285714285714e-05, |
|
"loss": 0.0898, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 39.375, |
|
"grad_norm": 0.486328125, |
|
"learning_rate": 2.4285714285714288e-05, |
|
"loss": 0.0898, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"grad_norm": 1.078125, |
|
"learning_rate": 2.2857142857142858e-05, |
|
"loss": 0.0931, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 40.625, |
|
"grad_norm": 0.404296875, |
|
"learning_rate": 2.1428571428571428e-05, |
|
"loss": 0.0901, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 41.25, |
|
"grad_norm": 0.640625, |
|
"learning_rate": 2e-05, |
|
"loss": 0.0931, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 41.875, |
|
"grad_norm": 0.6953125, |
|
"learning_rate": 1.8571428571428572e-05, |
|
"loss": 0.0921, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 42.5, |
|
"grad_norm": 0.671875, |
|
"learning_rate": 1.7142857142857145e-05, |
|
"loss": 0.091, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 43.125, |
|
"grad_norm": 0.671875, |
|
"learning_rate": 1.5714285714285715e-05, |
|
"loss": 0.0943, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 43.75, |
|
"grad_norm": 0.734375, |
|
"learning_rate": 1.4285714285714285e-05, |
|
"loss": 0.0933, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 44.375, |
|
"grad_norm": 0.80859375, |
|
"learning_rate": 1.2857142857142857e-05, |
|
"loss": 0.0907, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 45.0, |
|
"grad_norm": 0.7421875, |
|
"learning_rate": 1.1428571428571429e-05, |
|
"loss": 0.0907, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 45.625, |
|
"grad_norm": 0.97265625, |
|
"learning_rate": 1e-05, |
|
"loss": 0.0882, |
|
"step": 730 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 800, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 50, |
|
"save_steps": 10, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 2799985827465216.0, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|