-
Update README.md
-
Improved model: step 5750 (eval_loss=0.0073)
-
Improved model: step 5250 (eval_loss=0.0074)
-
Improved model: step 4250 (eval_loss=0.0076)
-
Improved model: step 4000 (eval_loss=0.0086)
-
Improved model: step 3500 (eval_loss=0.0089)
-
Improved model: step 3250 (eval_loss=0.0097)
-
Improved model: step 2750 (eval_loss=0.0106)
-
Improved model: step 2500 (eval_loss=0.0120)
-
Improved model: step 2250 (eval_loss=0.0146)
-
Improved model: step 2000 (eval_loss=0.0212)
-
Improved model: step 1750 (eval_loss=0.0465)
-
Improved model: step 1500 (eval_loss=0.0840)
-
Improved model: step 1250 (eval_loss=0.1190)
-
Improved model: step 1000 (eval_loss=0.2299)
-
Improved model: step 750 (eval_loss=0.2961)
-
Improved model: step 500 (eval_loss=0.4009)
-
Improved model: step 250 (eval_loss=0.7174)
-
Improved model: step 5750 (eval_loss=0.0072)
-
Improved model: step 5500 (eval_loss=0.0074)
-
Improved model: step 5250 (eval_loss=0.0074)
-
Improved model: step 4250 (eval_loss=0.0077)
-
Improved model: step 4000 (eval_loss=0.0087)
-
Improved model: step 3750 (eval_loss=0.0088)
-
Improved model: step 3500 (eval_loss=0.0089)
-
Improved model: step 3250 (eval_loss=0.0090)
-
Improved model: step 3000 (eval_loss=0.0102)
-
Improved model: step 2750 (eval_loss=0.0113)
-
Improved model: step 2500 (eval_loss=0.0131)
-
Improved model: step 2250 (eval_loss=0.0197)
-
Improved model: step 2000 (eval_loss=0.0354)
-
Improved model: step 1750 (eval_loss=0.0641)
-
Improved model: step 1500 (eval_loss=0.1526)
-
Improved model: step 1250 (eval_loss=0.1894)
-
Improved model: step 1000 (eval_loss=0.2543)
-
Improved model: step 750 (eval_loss=0.2869)
-
Improved model: step 500 (eval_loss=0.3959)
-
Improved model: step 250 (eval_loss=0.6936)
-
Improved model: step 6000 (eval_loss=0.0080)
-
Improved model: step 5750 (eval_loss=0.0081)
-
Improved model: step 5250 (eval_loss=0.0083)
-
Improved model: step 5000 (eval_loss=0.0084)
-
Improved model: step 4750 (eval_loss=0.0086)
-
Improved model: step 3750 (eval_loss=0.0090)
-
Improved model: step 3500 (eval_loss=0.0092)
-
Improved model: step 3250 (eval_loss=0.0098)
-
Improved model: step 3000 (eval_loss=0.0106)
-
Improved model: step 2750 (eval_loss=0.0107)
-
Improved model: step 2500 (eval_loss=0.0121)
-
Improved model: step 2250 (eval_loss=0.0134)