|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 10317, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.6129032258064517e-05, |
|
"loss": 1.0967, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.2258064516129034e-05, |
|
"loss": 0.1342, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.8387096774193554e-05, |
|
"loss": 0.1196, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.9990021664883965e-05, |
|
"loss": 0.1065, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.995553888081917e-05, |
|
"loss": 0.1011, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.989646243314479e-05, |
|
"loss": 0.1021, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.9812850541652764e-05, |
|
"loss": 0.0961, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.970478560579529e-05, |
|
"loss": 0.089, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.9572374123480235e-05, |
|
"loss": 0.0853, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.941574658611754e-05, |
|
"loss": 0.0866, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.923505735002e-05, |
|
"loss": 0.0824, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.90304844842853e-05, |
|
"loss": 0.0846, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.880222959530902e-05, |
|
"loss": 0.0837, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.85505176281017e-05, |
|
"loss": 0.0792, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.8275596644605725e-05, |
|
"loss": 0.0757, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.7977737579230465e-05, |
|
"loss": 0.0769, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.765723397184665e-05, |
|
"loss": 0.0725, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.731440167850305e-05, |
|
"loss": 0.0772, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.6949578560150574e-05, |
|
"loss": 0.0728, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.65631241496806e-05, |
|
"loss": 0.066, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.615541929760561e-05, |
|
"loss": 0.0715, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.572686579673125e-05, |
|
"loss": 0.0678, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.527788598618992e-05, |
|
"loss": 0.0663, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.480892233522584e-05, |
|
"loss": 0.0691, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.432043700714202e-05, |
|
"loss": 0.0688, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.381291140383861e-05, |
|
"loss": 0.0633, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.3286845691391785e-05, |
|
"loss": 0.0662, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.274275830714048e-05, |
|
"loss": 0.061, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.218118544876673e-05, |
|
"loss": 0.0627, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.160268054587338e-05, |
|
"loss": 0.0637, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.100781371457958e-05, |
|
"loss": 0.0643, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.039717119567185e-05, |
|
"loss": 0.0606, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.9771354776864214e-05, |
|
"loss": 0.0588, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.913098119973694e-05, |
|
"loss": 0.0619, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.8476681551938166e-05, |
|
"loss": 0.0584, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.7809100645247544e-05, |
|
"loss": 0.0531, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.7128896380114734e-05, |
|
"loss": 0.0501, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.643673909729906e-05, |
|
"loss": 0.0528, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.5733310917249186e-05, |
|
"loss": 0.0494, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.5019305067874016e-05, |
|
"loss": 0.056, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.4295425201367106e-05, |
|
"loss": 0.0484, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 1.22, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.356238470075799e-05, |
|
"loss": 0.0508, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.28209059768738e-05, |
|
"loss": 0.0466, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.207171975640394e-05, |
|
"loss": 0.0427, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.1315564361769544e-05, |
|
"loss": 0.0516, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.0553184983507316e-05, |
|
"loss": 0.0512, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.9785332945884835e-05, |
|
"loss": 0.0501, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.9012764966471118e-05, |
|
"loss": 0.0465, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.8236242410392038e-05, |
|
"loss": 0.0486, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.7456530540005604e-05, |
|
"loss": 0.049, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.6674397760736534e-05, |
|
"loss": 0.0446, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.589061486381331e-05, |
|
"loss": 0.0445, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.5105954266654043e-05, |
|
"loss": 0.043, |
|
"step": 5300 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.432118925164972e-05, |
|
"loss": 0.047, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.3537093204095038e-05, |
|
"loss": 0.0451, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.27544388500178e-05, |
|
"loss": 0.0432, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.197399749465805e-05, |
|
"loss": 0.0426, |
|
"step": 5700 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.1196538262347358e-05, |
|
"loss": 0.0433, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.0422827338537416e-05, |
|
"loss": 0.0451, |
|
"step": 5900 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.965362721472488e-05, |
|
"loss": 0.0476, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.888969593701663e-05, |
|
"loss": 0.0418, |
|
"step": 6100 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.8131786359075914e-05, |
|
"loss": 0.0424, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.738064540018567e-05, |
|
"loss": 0.0416, |
|
"step": 6300 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.6637013309160194e-05, |
|
"loss": 0.0436, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.5901622934830474e-05, |
|
"loss": 0.0419, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.5175199003822325e-05, |
|
"loss": 0.0405, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.445845740633881e-05, |
|
"loss": 0.0392, |
|
"step": 6700 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.3752104490651061e-05, |
|
"loss": 0.0383, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.305683636699259e-05, |
|
"loss": 0.0364, |
|
"step": 6900 |
|
}, |
|
{ |
|
"epoch": 2.04, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.237333822154326e-05, |
|
"loss": 0.0282, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.1702283641178782e-05, |
|
"loss": 0.028, |
|
"step": 7100 |
|
}, |
|
{ |
|
"epoch": 2.09, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.1044333949651476e-05, |
|
"loss": 0.0296, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 2.12, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.0400137555856137e-05, |
|
"loss": 0.0263, |
|
"step": 7300 |
|
}, |
|
{ |
|
"epoch": 2.15, |
|
"grad_norm": 0.0, |
|
"learning_rate": 9.770329314823701e-06, |
|
"loss": 0.028, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 2.18, |
|
"grad_norm": 0.0, |
|
"learning_rate": 9.15552990207201e-06, |
|
"loss": 0.0279, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"grad_norm": 0.0, |
|
"learning_rate": 8.556345201930674e-06, |
|
"loss": 0.0257, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"grad_norm": 0.0, |
|
"learning_rate": 7.97336571044251e-06, |
|
"loss": 0.028, |
|
"step": 7700 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"grad_norm": 0.0, |
|
"learning_rate": 7.4071659534301424e-06, |
|
"loss": 0.0271, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"grad_norm": 0.0, |
|
"learning_rate": 6.858303920301321e-06, |
|
"loss": 0.0254, |
|
"step": 7900 |
|
}, |
|
{ |
|
"epoch": 2.33, |
|
"grad_norm": 0.0, |
|
"learning_rate": 6.327320514150742e-06, |
|
"loss": 0.0252, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.36, |
|
"grad_norm": 0.0, |
|
"learning_rate": 5.8147390187005465e-06, |
|
"loss": 0.027, |
|
"step": 8100 |
|
}, |
|
{ |
|
"epoch": 2.38, |
|
"grad_norm": 0.0, |
|
"learning_rate": 5.3210645826045565e-06, |
|
"loss": 0.026, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.846783721624764e-06, |
|
"loss": 0.0286, |
|
"step": 8300 |
|
}, |
|
{ |
|
"epoch": 2.44, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.392363839170357e-06, |
|
"loss": 0.0245, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.9582527656720964e-06, |
|
"loss": 0.0256, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.5448783172457367e-06, |
|
"loss": 0.0253, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.152647874079628e-06, |
|
"loss": 0.0244, |
|
"step": 8700 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.781947978961827e-06, |
|
"loss": 0.0224, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 2.59, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.4331439563425234e-06, |
|
"loss": 0.0246, |
|
"step": 8900 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.106579552307053e-06, |
|
"loss": 0.0241, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.65, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.8025765958144453e-06, |
|
"loss": 0.0242, |
|
"step": 9100 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.5214346815351942e-06, |
|
"loss": 0.0258, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 2.7, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.2634308746009932e-06, |
|
"loss": 0.0226, |
|
"step": 9300 |
|
}, |
|
{ |
|
"epoch": 2.73, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.0288194375572436e-06, |
|
"loss": 0.0254, |
|
"step": 9400 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"grad_norm": 0.0, |
|
"learning_rate": 8.178315797875257e-07, |
|
"loss": 0.0236, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 2.79, |
|
"grad_norm": 0.0, |
|
"learning_rate": 6.30675229656949e-07, |
|
"loss": 0.0247, |
|
"step": 9600 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"grad_norm": 0.0, |
|
"learning_rate": 4.6753482959888406e-07, |
|
"loss": 0.0263, |
|
"step": 9700 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.285711543471265e-07, |
|
"loss": 0.0237, |
|
"step": 9800 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"grad_norm": 0.0, |
|
"learning_rate": 2.1392115249250222e-07, |
|
"loss": 0.0251, |
|
"step": 9900 |
|
}, |
|
{ |
|
"epoch": 2.91, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.2369781152015425e-07, |
|
"loss": 0.0253, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 2.94, |
|
"grad_norm": 0.0, |
|
"learning_rate": 5.799004646045436e-08, |
|
"loss": 0.0246, |
|
"step": 10100 |
|
}, |
|
{ |
|
"epoch": 2.97, |
|
"grad_norm": 0.0, |
|
"learning_rate": 1.6862612263324195e-08, |
|
"loss": 0.0247, |
|
"step": 10200 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 0.0, |
|
"learning_rate": 3.560399822716276e-10, |
|
"loss": 0.0241, |
|
"step": 10300 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 10317, |
|
"total_flos": 1.8160995702249882e+18, |
|
"train_loss": 0.059974256984898895, |
|
"train_runtime": 16482.6502, |
|
"train_samples_per_second": 2.503, |
|
"train_steps_per_second": 0.626 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 10317, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 1.8160995702249882e+18, |
|
"train_batch_size": 1, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|