artifacts/adamw_sweep_cmpatino-0/log_lr0.0020_wd0.05_steps2812_cmpatino-0.txt

W0430 19:58:52.581000 1965086 torch/distributed/run.py:851] 
W0430 19:58:52.581000 1965086 torch/distributed/run.py:851] *****************************************
W0430 19:58:52.581000 1965086 torch/distributed/run.py:851] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
W0430 19:58:52.581000 1965086 torch/distributed/run.py:851] *****************************************
logs/lr0.0020_wd0.05_steps2812_cmpatino-0_81669578-ab98-4c00-9611-3014a0c9b0f7.txt
CONFIG block_lr=0.002 block_wd=0.05 betas=(0.9,0.95) train_steps=2812 warmup=125 cooldown_frac=0.7
step:0/2812 val_loss:10.82585 train_time:0.000s step_avg:nanms
step:125/2812 val_loss:6.52563 train_time:70.471s step_avg:563.77ms
step:250/2812 val_loss:5.63860 train_time:138.750s step_avg:546.23ms
step:375/2812 val_loss:5.09376 train_time:206.845s step_avg:544.76ms
step:500/2812 val_loss:4.74756 train_time:274.805s step_avg:543.68ms
step:625/2812 val_loss:4.44901 train_time:342.720s step_avg:543.32ms
step:750/2812 val_loss:4.23819 train_time:410.563s step_avg:542.75ms
step:875/2812 val_loss:4.10279 train_time:478.407s step_avg:542.75ms
step:1000/2812 val_loss:3.99920 train_time:546.230s step_avg:542.59ms
step:1125/2812 val_loss:3.93021 train_time:614.106s step_avg:543.01ms
step:1250/2812 val_loss:3.86073 train_time:681.928s step_avg:542.58ms
step:1375/2812 val_loss:3.81861 train_time:749.753s step_avg:542.60ms
step:1500/2812 val_loss:3.76132 train_time:817.565s step_avg:542.49ms
step:1625/2812 val_loss:3.72627 train_time:885.358s step_avg:542.35ms
step:1750/2812 val_loss:3.69413 train_time:953.204s step_avg:542.77ms
step:1875/2812 val_loss:3.66039 train_time:1021.028s step_avg:542.59ms
step:2000/2812 val_loss:3.63399 train_time:1088.892s step_avg:542.91ms
step:2125/2812 val_loss:3.61006 train_time:1156.766s step_avg:542.99ms
step:2250/2812 val_loss:3.58753 train_time:1224.631s step_avg:542.92ms
step:2375/2812 val_loss:3.56775 train_time:1292.502s step_avg:542.97ms
step:2500/2812 val_loss:3.55013 train_time:1360.330s step_avg:542.62ms
step:2625/2812 val_loss:3.53476 train_time:1428.185s step_avg:542.84ms
step:2750/2812 val_loss:3.52333 train_time:1496.050s step_avg:542.92ms
step:2812/2812 val_loss:3.52063 train_time:1529.704s step_avg:542.81ms