| { | |
| "best_metric": 0.3883955776691437, | |
| "best_model_checkpoint": "exp/vicuna-7b-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": 1.4343, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 1.4848, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 1.1941, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.8226, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00015, | |
| "loss": 0.6671, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.5676, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.5655, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.5251, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00027, | |
| "loss": 0.4845, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0003, | |
| "loss": 0.481, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0002976377952755905, | |
| "loss": 0.4565, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002952755905511811, | |
| "loss": 0.4625, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00029291338582677163, | |
| "loss": 0.4584, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00029055118110236217, | |
| "loss": 0.4425, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.0002881889763779527, | |
| "loss": 0.4573, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002858267716535433, | |
| "loss": 0.4361, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00028346456692913383, | |
| "loss": 0.4396, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00028110236220472436, | |
| "loss": 0.4391, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00027874015748031495, | |
| "loss": 0.418, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002763779527559055, | |
| "loss": 0.4469, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 0.4269736409187317, | |
| "eval_runtime": 19.352, | |
| "eval_samples_per_second": 103.348, | |
| "eval_steps_per_second": 1.654, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002740157480314961, | |
| "loss": 0.4149, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00027165354330708656, | |
| "loss": 0.428, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00026929133858267715, | |
| "loss": 0.4248, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0002669291338582677, | |
| "loss": 0.4249, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0002645669291338582, | |
| "loss": 0.4331, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002622047244094488, | |
| "loss": 0.4192, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00025984251968503934, | |
| "loss": 0.4204, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00025748031496062993, | |
| "loss": 0.4318, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00025511811023622047, | |
| "loss": 0.4229, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.000252755905511811, | |
| "loss": 0.4214, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.00025039370078740154, | |
| "loss": 0.416, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00024803149606299207, | |
| "loss": 0.4199, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00024566929133858266, | |
| "loss": 0.4218, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002433070866141732, | |
| "loss": 0.4113, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00024094488188976376, | |
| "loss": 0.4185, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00023858267716535432, | |
| "loss": 0.4168, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00023622047244094488, | |
| "loss": 0.4162, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0002338582677165354, | |
| "loss": 0.4175, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00023149606299212595, | |
| "loss": 0.4045, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00022913385826771652, | |
| "loss": 0.4152, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 0.4086858630180359, | |
| "eval_runtime": 19.2818, | |
| "eval_samples_per_second": 103.725, | |
| "eval_steps_per_second": 1.66, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.00022677165354330705, | |
| "loss": 0.415, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00022440944881889761, | |
| "loss": 0.4091, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00022204724409448818, | |
| "loss": 0.4132, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00021968503937007874, | |
| "loss": 0.3985, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00021732283464566927, | |
| "loss": 0.4056, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002149606299212598, | |
| "loss": 0.4005, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00021259842519685037, | |
| "loss": 0.4059, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002102362204724409, | |
| "loss": 0.409, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00020787401574803147, | |
| "loss": 0.4031, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00020551181102362203, | |
| "loss": 0.4097, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.0002031496062992126, | |
| "loss": 0.4017, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00020078740157480313, | |
| "loss": 0.4026, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0001984251968503937, | |
| "loss": 0.4106, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.00019606299212598423, | |
| "loss": 0.395, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.0001937007874015748, | |
| "loss": 0.3988, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00019133858267716532, | |
| "loss": 0.409, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00018897637795275589, | |
| "loss": 0.3997, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.00018661417322834645, | |
| "loss": 0.4007, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.000184251968503937, | |
| "loss": 0.3905, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.00018188976377952755, | |
| "loss": 0.4005, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 0.40032637119293213, | |
| "eval_runtime": 19.2818, | |
| "eval_samples_per_second": 103.725, | |
| "eval_steps_per_second": 1.66, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0001795275590551181, | |
| "loss": 0.3983, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.00017716535433070864, | |
| "loss": 0.3881, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.00017480314960629918, | |
| "loss": 0.4008, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00017244094488188974, | |
| "loss": 0.3927, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0001700787401574803, | |
| "loss": 0.4005, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.00016771653543307086, | |
| "loss": 0.3962, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.0001653543307086614, | |
| "loss": 0.3902, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.00016299212598425196, | |
| "loss": 0.3911, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.00016062992125984252, | |
| "loss": 0.3891, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.00015826771653543303, | |
| "loss": 0.3939, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 0.0001559055118110236, | |
| "loss": 0.4001, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.00015354330708661416, | |
| "loss": 0.3918, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.00015118110236220472, | |
| "loss": 0.3979, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.00014881889763779525, | |
| "loss": 0.3793, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00014645669291338582, | |
| "loss": 0.3879, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.00014409448818897635, | |
| "loss": 0.3915, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 0.00014173228346456691, | |
| "loss": 0.3831, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00013937007874015748, | |
| "loss": 0.3838, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 0.00013700787401574804, | |
| "loss": 0.3734, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00013464566929133857, | |
| "loss": 0.3872, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 0.3944130539894104, | |
| "eval_runtime": 19.2596, | |
| "eval_samples_per_second": 103.844, | |
| "eval_steps_per_second": 1.662, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.0001322834645669291, | |
| "loss": 0.386, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00012992125984251967, | |
| "loss": 0.3799, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.00012755905511811023, | |
| "loss": 0.3895, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.00012519685039370077, | |
| "loss": 0.3852, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.00012283464566929133, | |
| "loss": 0.3879, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 0.00012047244094488188, | |
| "loss": 0.3892, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.00011811023622047244, | |
| "loss": 0.3801, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.00011574803149606298, | |
| "loss": 0.3802, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.00011338582677165353, | |
| "loss": 0.3863, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.00011102362204724409, | |
| "loss": 0.3792, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 0.00010866141732283464, | |
| "loss": 0.3923, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.00010629921259842519, | |
| "loss": 0.3753, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00010393700787401573, | |
| "loss": 0.3777, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.0001015748031496063, | |
| "loss": 0.3849, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 9.921259842519685e-05, | |
| "loss": 0.3775, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 9.68503937007874e-05, | |
| "loss": 0.3853, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 9.448818897637794e-05, | |
| "loss": 0.3719, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 9.21259842519685e-05, | |
| "loss": 0.3779, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 8.976377952755905e-05, | |
| "loss": 0.3921, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 8.740157480314959e-05, | |
| "loss": 0.3776, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "eval_loss": 0.3908761739730835, | |
| "eval_runtime": 19.2678, | |
| "eval_samples_per_second": 103.8, | |
| "eval_steps_per_second": 1.661, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 8.503937007874015e-05, | |
| "loss": 0.3889, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 8.26771653543307e-05, | |
| "loss": 0.3819, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 8.031496062992126e-05, | |
| "loss": 0.3758, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 7.79527559055118e-05, | |
| "loss": 0.3753, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 7.559055118110236e-05, | |
| "loss": 0.3737, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 7.322834645669291e-05, | |
| "loss": 0.3833, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 7.086614173228346e-05, | |
| "loss": 0.3625, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 6.850393700787402e-05, | |
| "loss": 0.3809, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 6.614173228346455e-05, | |
| "loss": 0.3751, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 6.377952755905512e-05, | |
| "loss": 0.3776, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 6.141732283464567e-05, | |
| "loss": 0.3748, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 5.905511811023622e-05, | |
| "loss": 0.3636, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 5.669291338582676e-05, | |
| "loss": 0.372, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 5.433070866141732e-05, | |
| "loss": 0.3795, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 5.196850393700787e-05, | |
| "loss": 0.3632, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.960629921259842e-05, | |
| "loss": 0.3806, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 4.724409448818897e-05, | |
| "loss": 0.3732, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 4.488188976377953e-05, | |
| "loss": 0.3818, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 4.2519685039370076e-05, | |
| "loss": 0.3766, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 4.015748031496063e-05, | |
| "loss": 0.3587, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "eval_loss": 0.3883955776691437, | |
| "eval_runtime": 19.3219, | |
| "eval_samples_per_second": 103.51, | |
| "eval_steps_per_second": 1.656, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1370, | |
| "num_train_epochs": 5, | |
| "save_steps": 200, | |
| "total_flos": 2.2400975031249142e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |