|
{ |
|
"best_metric": 0.3790667653083801, |
|
"best_model_checkpoint": "exp/vicuna-13b-lora-sft-code_qa_desc_summ_triplet_r_16_alpha_32_8GPUs-0116/checkpoint-1200", |
|
"epoch": 4.375569735642662, |
|
"eval_steps": 200, |
|
"global_step": 1200, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 2.9999999999999997e-05, |
|
"loss": 1.374, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 5.9999999999999995e-05, |
|
"loss": 1.416, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 8.999999999999999e-05, |
|
"loss": 1.0841, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 0.00011999999999999999, |
|
"loss": 0.6951, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 0.00015, |
|
"loss": 0.6106, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.00017999999999999998, |
|
"loss": 0.5182, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 0.00020999999999999998, |
|
"loss": 0.5251, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 0.00023999999999999998, |
|
"loss": 0.4988, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 0.00027, |
|
"loss": 0.4838, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 0.0003, |
|
"loss": 0.4929, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 0.0002976377952755905, |
|
"loss": 0.4463, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 0.0002952755905511811, |
|
"loss": 0.4402, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 0.00029291338582677163, |
|
"loss": 0.4414, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 0.00029055118110236217, |
|
"loss": 0.4278, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 0.0002881889763779527, |
|
"loss": 0.4346, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 0.0002858267716535433, |
|
"loss": 0.4163, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 0.00028346456692913383, |
|
"loss": 0.4221, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 0.00028110236220472436, |
|
"loss": 0.4257, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 0.00027874015748031495, |
|
"loss": 0.4166, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 0.0002763779527559055, |
|
"loss": 0.4244, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"eval_loss": 0.4145399034023285, |
|
"eval_runtime": 30.3529, |
|
"eval_samples_per_second": 65.892, |
|
"eval_steps_per_second": 1.054, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 0.0002740157480314961, |
|
"loss": 0.407, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 0.00027165354330708656, |
|
"loss": 0.4103, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 0.00026929133858267715, |
|
"loss": 0.4116, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 0.0002669291338582677, |
|
"loss": 0.4027, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 0.0002645669291338582, |
|
"loss": 0.4217, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 0.0002622047244094488, |
|
"loss": 0.3988, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 0.00025984251968503934, |
|
"loss": 0.4146, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 0.00025748031496062993, |
|
"loss": 0.4155, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 0.00025511811023622047, |
|
"loss": 0.4022, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 0.000252755905511811, |
|
"loss": 0.4044, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 0.00025039370078740154, |
|
"loss": 0.4001, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 0.00024803149606299207, |
|
"loss": 0.3979, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 0.00024566929133858266, |
|
"loss": 0.405, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 0.0002433070866141732, |
|
"loss": 0.3959, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 0.00024094488188976376, |
|
"loss": 0.4061, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 0.00023858267716535432, |
|
"loss": 0.3914, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 0.00023622047244094488, |
|
"loss": 0.3978, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 0.0002338582677165354, |
|
"loss": 0.4008, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 0.00023149606299212595, |
|
"loss": 0.3908, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 0.00022913385826771652, |
|
"loss": 0.4006, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"eval_loss": 0.398809552192688, |
|
"eval_runtime": 30.3867, |
|
"eval_samples_per_second": 65.818, |
|
"eval_steps_per_second": 1.053, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 0.00022677165354330705, |
|
"loss": 0.3892, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 0.00022440944881889761, |
|
"loss": 0.3955, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 0.00022204724409448818, |
|
"loss": 0.3891, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 0.00021968503937007874, |
|
"loss": 0.3867, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 0.00021732283464566927, |
|
"loss": 0.3917, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 0.0002149606299212598, |
|
"loss": 0.3808, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 0.00021259842519685037, |
|
"loss": 0.3868, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 0.0002102362204724409, |
|
"loss": 0.395, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 0.00020787401574803147, |
|
"loss": 0.3739, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"learning_rate": 0.00020551181102362203, |
|
"loss": 0.3859, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"learning_rate": 0.0002031496062992126, |
|
"loss": 0.3882, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"learning_rate": 0.00020078740157480313, |
|
"loss": 0.3816, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"learning_rate": 0.0001984251968503937, |
|
"loss": 0.3943, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 1.97, |
|
"learning_rate": 0.00019606299212598423, |
|
"loss": 0.3828, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 0.0001937007874015748, |
|
"loss": 0.3927, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"learning_rate": 0.00019133858267716532, |
|
"loss": 0.3916, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 0.00018897637795275589, |
|
"loss": 0.3769, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"learning_rate": 0.00018661417322834645, |
|
"loss": 0.3858, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"learning_rate": 0.000184251968503937, |
|
"loss": 0.3803, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"learning_rate": 0.00018188976377952755, |
|
"loss": 0.383, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.19, |
|
"eval_loss": 0.39037391543388367, |
|
"eval_runtime": 30.346, |
|
"eval_samples_per_second": 65.907, |
|
"eval_steps_per_second": 1.055, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.22, |
|
"learning_rate": 0.0001795275590551181, |
|
"loss": 0.3887, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"learning_rate": 0.00017716535433070864, |
|
"loss": 0.3715, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"learning_rate": 0.00017480314960629918, |
|
"loss": 0.3852, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"learning_rate": 0.00017244094488188974, |
|
"loss": 0.3744, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 0.0001700787401574803, |
|
"loss": 0.3903, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"learning_rate": 0.00016771653543307086, |
|
"loss": 0.3829, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 2.44, |
|
"learning_rate": 0.0001653543307086614, |
|
"loss": 0.3684, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 2.48, |
|
"learning_rate": 0.00016299212598425196, |
|
"loss": 0.3819, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"learning_rate": 0.00016062992125984252, |
|
"loss": 0.3743, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 2.55, |
|
"learning_rate": 0.00015826771653543303, |
|
"loss": 0.3763, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.59, |
|
"learning_rate": 0.0001559055118110236, |
|
"loss": 0.3805, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 2.63, |
|
"learning_rate": 0.00015354330708661416, |
|
"loss": 0.3734, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 2.66, |
|
"learning_rate": 0.00015118110236220472, |
|
"loss": 0.3808, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"learning_rate": 0.00014881889763779525, |
|
"loss": 0.3675, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 2.73, |
|
"learning_rate": 0.00014645669291338582, |
|
"loss": 0.3671, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 2.77, |
|
"learning_rate": 0.00014409448818897635, |
|
"loss": 0.3686, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 2.81, |
|
"learning_rate": 0.00014173228346456691, |
|
"loss": 0.3631, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 0.00013937007874015748, |
|
"loss": 0.369, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"learning_rate": 0.00013700787401574804, |
|
"loss": 0.3604, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 2.92, |
|
"learning_rate": 0.00013464566929133857, |
|
"loss": 0.3657, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.92, |
|
"eval_loss": 0.3851480185985565, |
|
"eval_runtime": 30.443, |
|
"eval_samples_per_second": 65.697, |
|
"eval_steps_per_second": 1.051, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.95, |
|
"learning_rate": 0.0001322834645669291, |
|
"loss": 0.3618, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"learning_rate": 0.00012992125984251967, |
|
"loss": 0.3723, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 0.00012755905511811023, |
|
"loss": 0.3665, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 3.06, |
|
"learning_rate": 0.00012519685039370077, |
|
"loss": 0.3738, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"learning_rate": 0.00012283464566929133, |
|
"loss": 0.3669, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 0.00012047244094488188, |
|
"loss": 0.3683, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 0.00011811023622047244, |
|
"loss": 0.3531, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 3.21, |
|
"learning_rate": 0.00011574803149606298, |
|
"loss": 0.368, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 3.25, |
|
"learning_rate": 0.00011338582677165353, |
|
"loss": 0.367, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 3.28, |
|
"learning_rate": 0.00011102362204724409, |
|
"loss": 0.3628, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 3.32, |
|
"learning_rate": 0.00010866141732283464, |
|
"loss": 0.3715, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 3.35, |
|
"learning_rate": 0.00010629921259842519, |
|
"loss": 0.3553, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 3.39, |
|
"learning_rate": 0.00010393700787401573, |
|
"loss": 0.3806, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 3.43, |
|
"learning_rate": 0.0001015748031496063, |
|
"loss": 0.3651, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 3.46, |
|
"learning_rate": 9.921259842519685e-05, |
|
"loss": 0.3537, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 3.5, |
|
"learning_rate": 9.68503937007874e-05, |
|
"loss": 0.3694, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 3.54, |
|
"learning_rate": 9.448818897637794e-05, |
|
"loss": 0.3569, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 3.57, |
|
"learning_rate": 9.21259842519685e-05, |
|
"loss": 0.3589, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 3.61, |
|
"learning_rate": 8.976377952755905e-05, |
|
"loss": 0.3685, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 3.65, |
|
"learning_rate": 8.740157480314959e-05, |
|
"loss": 0.3709, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.65, |
|
"eval_loss": 0.3818259835243225, |
|
"eval_runtime": 30.3549, |
|
"eval_samples_per_second": 65.887, |
|
"eval_steps_per_second": 1.054, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.68, |
|
"learning_rate": 8.503937007874015e-05, |
|
"loss": 0.3608, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 3.72, |
|
"learning_rate": 8.26771653543307e-05, |
|
"loss": 0.3591, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 3.76, |
|
"learning_rate": 8.031496062992126e-05, |
|
"loss": 0.3599, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 3.79, |
|
"learning_rate": 7.79527559055118e-05, |
|
"loss": 0.3601, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 3.83, |
|
"learning_rate": 7.559055118110236e-05, |
|
"loss": 0.3546, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 3.87, |
|
"learning_rate": 7.322834645669291e-05, |
|
"loss": 0.3629, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 3.9, |
|
"learning_rate": 7.086614173228346e-05, |
|
"loss": 0.3567, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 3.94, |
|
"learning_rate": 6.850393700787402e-05, |
|
"loss": 0.3514, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 3.97, |
|
"learning_rate": 6.614173228346455e-05, |
|
"loss": 0.356, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 6.377952755905512e-05, |
|
"loss": 0.3606, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 4.05, |
|
"learning_rate": 6.141732283464567e-05, |
|
"loss": 0.3586, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 4.08, |
|
"learning_rate": 5.905511811023622e-05, |
|
"loss": 0.3528, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 4.12, |
|
"learning_rate": 5.669291338582676e-05, |
|
"loss": 0.3472, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"learning_rate": 5.433070866141732e-05, |
|
"loss": 0.3612, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"learning_rate": 5.196850393700787e-05, |
|
"loss": 0.3482, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 4.23, |
|
"learning_rate": 4.960629921259842e-05, |
|
"loss": 0.3538, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 4.27, |
|
"learning_rate": 4.724409448818897e-05, |
|
"loss": 0.3588, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 4.3, |
|
"learning_rate": 4.488188976377953e-05, |
|
"loss": 0.357, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 4.34, |
|
"learning_rate": 4.2519685039370076e-05, |
|
"loss": 0.3615, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 4.38, |
|
"learning_rate": 4.015748031496063e-05, |
|
"loss": 0.3485, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 4.38, |
|
"eval_loss": 0.3790667653083801, |
|
"eval_runtime": 30.2759, |
|
"eval_samples_per_second": 66.059, |
|
"eval_steps_per_second": 1.057, |
|
"step": 1200 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 1370, |
|
"num_train_epochs": 5, |
|
"save_steps": 200, |
|
"total_flos": 4.3556106687039406e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|