|
{ |
|
"best_metric": 1.2265406847000122, |
|
"best_model_checkpoint": "../llama2-9439-21sept/checkpoint-1400", |
|
"epoch": 2.431259044862518, |
|
"eval_steps": 200, |
|
"global_step": 1400, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 5.9999999999999995e-05, |
|
"loss": 3.4609, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 0.00011999999999999999, |
|
"loss": 2.8346, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 0.00017999999999999998, |
|
"loss": 1.7597, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 0.00023999999999999998, |
|
"loss": 1.542, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 0.0003, |
|
"loss": 1.4567, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 0.0002963076923076923, |
|
"loss": 1.4081, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 0.0002926153846153846, |
|
"loss": 1.3805, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 0.0002889230769230769, |
|
"loss": 1.3605, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 0.00028523076923076923, |
|
"loss": 1.3329, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 0.0002815384615384615, |
|
"loss": 1.3367, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"eval_loss": 1.3450770378112793, |
|
"eval_runtime": 6.306, |
|
"eval_samples_per_second": 47.574, |
|
"eval_steps_per_second": 1.269, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 0.0002778461538461538, |
|
"loss": 1.3152, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 0.0002741538461538461, |
|
"loss": 1.312, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 0.00027046153846153843, |
|
"loss": 1.2883, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 0.00026676923076923074, |
|
"loss": 1.2843, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 0.00026307692307692306, |
|
"loss": 1.2609, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 0.00025938461538461537, |
|
"loss": 1.2707, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 0.0002556923076923077, |
|
"loss": 1.2624, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 0.00025199999999999995, |
|
"loss": 1.273, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 0.0002483076923076923, |
|
"loss": 1.251, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 0.0002446153846153846, |
|
"loss": 1.2648, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"eval_loss": 1.292348861694336, |
|
"eval_runtime": 6.304, |
|
"eval_samples_per_second": 47.589, |
|
"eval_steps_per_second": 1.269, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 0.0002409230769230769, |
|
"loss": 1.265, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 0.0002372307692307692, |
|
"loss": 1.2516, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 0.00023353846153846151, |
|
"loss": 1.2441, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 0.00022984615384615383, |
|
"loss": 1.2204, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 0.00022615384615384614, |
|
"loss": 1.2221, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 0.00022246153846153846, |
|
"loss": 1.2246, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 0.00021876923076923074, |
|
"loss": 1.2227, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 0.00021507692307692306, |
|
"loss": 1.2124, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 0.00021138461538461537, |
|
"loss": 1.2065, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"learning_rate": 0.00020769230769230766, |
|
"loss": 1.2106, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"eval_loss": 1.2669302225112915, |
|
"eval_runtime": 6.3128, |
|
"eval_samples_per_second": 47.523, |
|
"eval_steps_per_second": 1.267, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 0.000204, |
|
"loss": 1.2046, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 1.11, |
|
"learning_rate": 0.00020030769230769229, |
|
"loss": 1.2114, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 0.0001966153846153846, |
|
"loss": 1.2122, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 0.00019310769230769227, |
|
"loss": 1.1963, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 1.22, |
|
"learning_rate": 0.00018941538461538461, |
|
"loss": 1.1965, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 0.0001857230769230769, |
|
"loss": 1.1937, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 0.00018203076923076921, |
|
"loss": 1.186, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 0.00017833846153846153, |
|
"loss": 1.1732, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 0.00017464615384615381, |
|
"loss": 1.1911, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 0.00017095384615384616, |
|
"loss": 1.2153, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"eval_loss": 1.2513903379440308, |
|
"eval_runtime": 6.3069, |
|
"eval_samples_per_second": 47.567, |
|
"eval_steps_per_second": 1.268, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 0.00016726153846153844, |
|
"loss": 1.1991, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 0.00016356923076923073, |
|
"loss": 1.1934, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 0.00015987692307692307, |
|
"loss": 1.2047, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 0.00015618461538461536, |
|
"loss": 1.2204, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 0.0001524923076923077, |
|
"loss": 1.1785, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 0.00014879999999999998, |
|
"loss": 1.1756, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"learning_rate": 0.0001451076923076923, |
|
"loss": 1.214, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 0.0001414153846153846, |
|
"loss": 1.1794, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"learning_rate": 0.00013772307692307693, |
|
"loss": 1.1946, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 0.0001340307692307692, |
|
"loss": 1.175, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"eval_loss": 1.2406948804855347, |
|
"eval_runtime": 6.3315, |
|
"eval_samples_per_second": 47.382, |
|
"eval_steps_per_second": 1.264, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 0.00013033846153846153, |
|
"loss": 1.178, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"learning_rate": 0.00012664615384615384, |
|
"loss": 1.1859, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 0.00012295384615384615, |
|
"loss": 1.1801, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"learning_rate": 0.00011926153846153845, |
|
"loss": 1.1743, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"learning_rate": 0.00011556923076923076, |
|
"loss": 1.1596, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"learning_rate": 0.00011187692307692307, |
|
"loss": 1.1495, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"learning_rate": 0.00010818461538461537, |
|
"loss": 1.1808, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 0.00010449230769230768, |
|
"loss": 1.1743, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"learning_rate": 0.0001008, |
|
"loss": 1.1354, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 9.71076923076923e-05, |
|
"loss": 1.1712, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"eval_loss": 1.2316911220550537, |
|
"eval_runtime": 6.3191, |
|
"eval_samples_per_second": 47.475, |
|
"eval_steps_per_second": 1.266, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"learning_rate": 9.34153846153846e-05, |
|
"loss": 1.1709, |
|
"step": 1220 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"learning_rate": 8.972307692307691e-05, |
|
"loss": 1.1667, |
|
"step": 1240 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"learning_rate": 8.603076923076923e-05, |
|
"loss": 1.1781, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 2.22, |
|
"learning_rate": 8.233846153846154e-05, |
|
"loss": 1.1628, |
|
"step": 1280 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"learning_rate": 7.864615384615383e-05, |
|
"loss": 1.1723, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 2.29, |
|
"learning_rate": 7.495384615384615e-05, |
|
"loss": 1.1578, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"learning_rate": 7.126153846153845e-05, |
|
"loss": 1.1399, |
|
"step": 1340 |
|
}, |
|
{ |
|
"epoch": 2.36, |
|
"learning_rate": 6.756923076923077e-05, |
|
"loss": 1.1612, |
|
"step": 1360 |
|
}, |
|
{ |
|
"epoch": 2.4, |
|
"learning_rate": 6.387692307692307e-05, |
|
"loss": 1.1582, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"learning_rate": 6.0184615384615375e-05, |
|
"loss": 1.1534, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"eval_loss": 1.2265406847000122, |
|
"eval_runtime": 6.3176, |
|
"eval_samples_per_second": 47.486, |
|
"eval_steps_per_second": 1.266, |
|
"step": 1400 |
|
} |
|
], |
|
"logging_steps": 20, |
|
"max_steps": 1725, |
|
"num_train_epochs": 3, |
|
"save_steps": 200, |
|
"total_flos": 3.348909438051287e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|