|
{ |
|
"best_metric": NaN, |
|
"best_model_checkpoint": "/mnt/bn/qingyi-bn-lq/llama/saved_bloomz-7b1-mt_TQA/checkpoint-200", |
|
"epoch": 2.7962861824139815, |
|
"global_step": 800, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 5.9999999999999995e-05, |
|
"loss": 56.1213, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 0.00011999999999999999, |
|
"loss": 46.3785, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 0.00017999999999999998, |
|
"loss": 27.2008, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 0.00023999999999999998, |
|
"loss": 58.3372, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 0.0003, |
|
"loss": 21.5097, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 0.00029208443271767806, |
|
"loss": 149.5568, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 0.0002841688654353562, |
|
"loss": 12.1125, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 0.0002762532981530343, |
|
"loss": 64.6813, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 0.00026833773087071237, |
|
"loss": 12.8244, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 0.0002604221635883905, |
|
"loss": 13.6392, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"eval_loss": NaN, |
|
"eval_runtime": 39.2226, |
|
"eval_samples_per_second": 50.991, |
|
"eval_steps_per_second": 1.606, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 0.0002525065963060686, |
|
"loss": 29.1801, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 0.0002445910290237467, |
|
"loss": 7.1808, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 0.0002366754617414248, |
|
"loss": 17.5066, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 0.00022875989445910288, |
|
"loss": 13.7954, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 0.000220844327176781, |
|
"loss": 15.2006, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 0.00021292875989445908, |
|
"loss": 18.5282, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 0.0002050131926121372, |
|
"loss": 10.157, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"learning_rate": 0.0001970976253298153, |
|
"loss": 9.2541, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 0.00018918205804749339, |
|
"loss": 9.0145, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"learning_rate": 0.00018126649076517147, |
|
"loss": 8.6491, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"eval_loss": NaN, |
|
"eval_runtime": 39.8237, |
|
"eval_samples_per_second": 50.221, |
|
"eval_steps_per_second": 1.582, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 0.0001733509234828496, |
|
"loss": 6.757, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 0.0001654353562005277, |
|
"loss": 4.1081, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 0.00015751978891820578, |
|
"loss": 19.5577, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 0.0001496042216358839, |
|
"loss": 14.4416, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 0.00014168865435356198, |
|
"loss": 8.4368, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"learning_rate": 0.0001337730870712401, |
|
"loss": 5.4915, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 0.00012585751978891818, |
|
"loss": 28.9042, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"learning_rate": 0.00011794195250659629, |
|
"loss": 25.3276, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 0.0001100263852242744, |
|
"loss": 8.954, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 0.0001021108179419525, |
|
"loss": 8.7963, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_loss": NaN, |
|
"eval_runtime": 39.412, |
|
"eval_samples_per_second": 50.746, |
|
"eval_steps_per_second": 1.598, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 9.41952506596306e-05, |
|
"loss": 17.8767, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"learning_rate": 8.62796833773087e-05, |
|
"loss": 19.0455, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"learning_rate": 7.836411609498681e-05, |
|
"loss": 6.1526, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 2.38, |
|
"learning_rate": 7.044854881266491e-05, |
|
"loss": 8.0983, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 2.45, |
|
"learning_rate": 6.2532981530343e-05, |
|
"loss": 7.2175, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"learning_rate": 5.461741424802111e-05, |
|
"loss": 12.3619, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 2.59, |
|
"learning_rate": 4.670184696569921e-05, |
|
"loss": 6.1864, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 3.8786279683377306e-05, |
|
"loss": 6.9226, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 2.73, |
|
"learning_rate": 3.0870712401055405e-05, |
|
"loss": 7.2459, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"learning_rate": 2.2955145118733507e-05, |
|
"loss": 7.1272, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.8, |
|
"eval_loss": NaN, |
|
"eval_runtime": 39.4817, |
|
"eval_samples_per_second": 50.656, |
|
"eval_steps_per_second": 1.596, |
|
"step": 800 |
|
} |
|
], |
|
"max_steps": 858, |
|
"num_train_epochs": 3, |
|
"total_flos": 7.546667811527983e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|