End of training
Browse files- trainer_state.json +1618 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1618 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.17242242242242242,
|
| 3 |
+
"best_model_checkpoint": "w2v2_ablation_with_ling_head-0drop-load-best-per-best_on_tp0.025_tl10_fp0.001_fl16/checkpoint-3000",
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 10600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.94,
|
| 13 |
+
"learning_rate": 1.3396226415094341e-06,
|
| 14 |
+
"loss": 114.032,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.94,
|
| 19 |
+
"eval_loss": 98.52102661132812,
|
| 20 |
+
"eval_runtime": 2.516,
|
| 21 |
+
"eval_samples_per_second": 284.973,
|
| 22 |
+
"eval_steps_per_second": 2.385,
|
| 23 |
+
"eval_wer": 21.496663329996665,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.89,
|
| 28 |
+
"learning_rate": 3.2264150943396226e-06,
|
| 29 |
+
"loss": 63.4912,
|
| 30 |
+
"step": 200
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.89,
|
| 34 |
+
"eval_loss": 9.89896011352539,
|
| 35 |
+
"eval_runtime": 1.6869,
|
| 36 |
+
"eval_samples_per_second": 425.031,
|
| 37 |
+
"eval_steps_per_second": 3.557,
|
| 38 |
+
"eval_wer": 1.0,
|
| 39 |
+
"step": 200
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.83,
|
| 43 |
+
"learning_rate": 5.1132075471698114e-06,
|
| 44 |
+
"loss": 5.9426,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.83,
|
| 49 |
+
"eval_loss": 5.290933132171631,
|
| 50 |
+
"eval_runtime": 1.6725,
|
| 51 |
+
"eval_samples_per_second": 428.701,
|
| 52 |
+
"eval_steps_per_second": 3.587,
|
| 53 |
+
"eval_wer": 1.0,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.77,
|
| 58 |
+
"learning_rate": 7e-06,
|
| 59 |
+
"loss": 5.0183,
|
| 60 |
+
"step": 400
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 3.77,
|
| 64 |
+
"eval_loss": 5.248229503631592,
|
| 65 |
+
"eval_runtime": 1.8648,
|
| 66 |
+
"eval_samples_per_second": 384.484,
|
| 67 |
+
"eval_steps_per_second": 3.217,
|
| 68 |
+
"eval_wer": 1.0,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 4.72,
|
| 73 |
+
"learning_rate": 8.88679245283019e-06,
|
| 74 |
+
"loss": 4.6782,
|
| 75 |
+
"step": 500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 4.72,
|
| 79 |
+
"eval_loss": 5.53143835067749,
|
| 80 |
+
"eval_runtime": 1.7224,
|
| 81 |
+
"eval_samples_per_second": 416.275,
|
| 82 |
+
"eval_steps_per_second": 3.483,
|
| 83 |
+
"eval_wer": 1.0,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.66,
|
| 88 |
+
"learning_rate": 1.077358490566038e-05,
|
| 89 |
+
"loss": 4.4732,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 5.66,
|
| 94 |
+
"eval_loss": 5.225007057189941,
|
| 95 |
+
"eval_runtime": 1.775,
|
| 96 |
+
"eval_samples_per_second": 403.947,
|
| 97 |
+
"eval_steps_per_second": 3.38,
|
| 98 |
+
"eval_wer": 1.0,
|
| 99 |
+
"step": 600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.6,
|
| 103 |
+
"learning_rate": 1.2660377358490568e-05,
|
| 104 |
+
"loss": 4.4059,
|
| 105 |
+
"step": 700
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.6,
|
| 109 |
+
"eval_loss": 5.148309230804443,
|
| 110 |
+
"eval_runtime": 1.6042,
|
| 111 |
+
"eval_samples_per_second": 446.945,
|
| 112 |
+
"eval_steps_per_second": 3.74,
|
| 113 |
+
"eval_wer": 1.0,
|
| 114 |
+
"step": 700
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.55,
|
| 118 |
+
"learning_rate": 1.4547169811320756e-05,
|
| 119 |
+
"loss": 4.3368,
|
| 120 |
+
"step": 800
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.55,
|
| 124 |
+
"eval_loss": 4.927549362182617,
|
| 125 |
+
"eval_runtime": 1.7761,
|
| 126 |
+
"eval_samples_per_second": 403.7,
|
| 127 |
+
"eval_steps_per_second": 3.378,
|
| 128 |
+
"eval_wer": 1.0,
|
| 129 |
+
"step": 800
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 8.49,
|
| 133 |
+
"learning_rate": 1.6433962264150943e-05,
|
| 134 |
+
"loss": 4.2178,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.49,
|
| 139 |
+
"eval_loss": 4.898721694946289,
|
| 140 |
+
"eval_runtime": 1.7092,
|
| 141 |
+
"eval_samples_per_second": 419.491,
|
| 142 |
+
"eval_steps_per_second": 3.51,
|
| 143 |
+
"eval_wer": 1.0,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.43,
|
| 148 |
+
"learning_rate": 1.832075471698113e-05,
|
| 149 |
+
"loss": 3.913,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.43,
|
| 154 |
+
"eval_loss": 3.700657606124878,
|
| 155 |
+
"eval_runtime": 1.878,
|
| 156 |
+
"eval_samples_per_second": 381.786,
|
| 157 |
+
"eval_steps_per_second": 3.195,
|
| 158 |
+
"eval_wer": 0.880714047380714,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 10.38,
|
| 163 |
+
"learning_rate": 1.99999343918343e-05,
|
| 164 |
+
"loss": 2.7998,
|
| 165 |
+
"step": 1100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 10.38,
|
| 169 |
+
"eval_loss": 2.129572629928589,
|
| 170 |
+
"eval_runtime": 1.7878,
|
| 171 |
+
"eval_samples_per_second": 401.044,
|
| 172 |
+
"eval_steps_per_second": 3.356,
|
| 173 |
+
"eval_wer": 0.5331164497831165,
|
| 174 |
+
"step": 1100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 11.32,
|
| 178 |
+
"learning_rate": 1.999332009015375e-05,
|
| 179 |
+
"loss": 1.8405,
|
| 180 |
+
"step": 1200
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 11.32,
|
| 184 |
+
"eval_loss": 1.487300992012024,
|
| 185 |
+
"eval_runtime": 1.6519,
|
| 186 |
+
"eval_samples_per_second": 434.046,
|
| 187 |
+
"eval_steps_per_second": 3.632,
|
| 188 |
+
"eval_wer": 0.4636302969636303,
|
| 189 |
+
"step": 1200
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 12.26,
|
| 193 |
+
"learning_rate": 1.9975869674932303e-05,
|
| 194 |
+
"loss": 1.2987,
|
| 195 |
+
"step": 1300
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.26,
|
| 199 |
+
"eval_loss": 1.0532265901565552,
|
| 200 |
+
"eval_runtime": 1.65,
|
| 201 |
+
"eval_samples_per_second": 434.557,
|
| 202 |
+
"eval_steps_per_second": 3.636,
|
| 203 |
+
"eval_wer": 0.3337504170837504,
|
| 204 |
+
"step": 1300
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 13.21,
|
| 208 |
+
"learning_rate": 1.9947602068277822e-05,
|
| 209 |
+
"loss": 1.0387,
|
| 210 |
+
"step": 1400
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.21,
|
| 214 |
+
"eval_loss": 0.8758745193481445,
|
| 215 |
+
"eval_runtime": 1.8254,
|
| 216 |
+
"eval_samples_per_second": 392.793,
|
| 217 |
+
"eval_steps_per_second": 3.287,
|
| 218 |
+
"eval_wer": 0.3347514180847514,
|
| 219 |
+
"step": 1400
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 14.15,
|
| 223 |
+
"learning_rate": 1.9908547921764814e-05,
|
| 224 |
+
"loss": 0.851,
|
| 225 |
+
"step": 1500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 14.15,
|
| 229 |
+
"eval_loss": 0.7742961645126343,
|
| 230 |
+
"eval_runtime": 1.8016,
|
| 231 |
+
"eval_samples_per_second": 397.975,
|
| 232 |
+
"eval_steps_per_second": 3.33,
|
| 233 |
+
"eval_wer": 0.36036036036036034,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 15.09,
|
| 238 |
+
"learning_rate": 1.985874958319781e-05,
|
| 239 |
+
"loss": 0.7128,
|
| 240 |
+
"step": 1600
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.09,
|
| 244 |
+
"eval_loss": 0.6522524952888489,
|
| 245 |
+
"eval_runtime": 1.8083,
|
| 246 |
+
"eval_samples_per_second": 396.509,
|
| 247 |
+
"eval_steps_per_second": 3.318,
|
| 248 |
+
"eval_wer": 0.279612946279613,
|
| 249 |
+
"step": 1600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.04,
|
| 253 |
+
"learning_rate": 1.9798261050692154e-05,
|
| 254 |
+
"loss": 0.605,
|
| 255 |
+
"step": 1700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 16.04,
|
| 259 |
+
"eval_loss": 0.6352096199989319,
|
| 260 |
+
"eval_runtime": 1.7571,
|
| 261 |
+
"eval_samples_per_second": 408.049,
|
| 262 |
+
"eval_steps_per_second": 3.415,
|
| 263 |
+
"eval_wer": 0.29946613279946616,
|
| 264 |
+
"step": 1700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.98,
|
| 268 |
+
"learning_rate": 1.9727147914121905e-05,
|
| 269 |
+
"loss": 0.5315,
|
| 270 |
+
"step": 1800
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.98,
|
| 274 |
+
"eval_loss": 0.591980516910553,
|
| 275 |
+
"eval_runtime": 1.9144,
|
| 276 |
+
"eval_samples_per_second": 374.536,
|
| 277 |
+
"eval_steps_per_second": 3.134,
|
| 278 |
+
"eval_wer": 0.2602602602602603,
|
| 279 |
+
"step": 1800
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 17.92,
|
| 283 |
+
"learning_rate": 1.96454872839984e-05,
|
| 284 |
+
"loss": 0.4845,
|
| 285 |
+
"step": 1900
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 17.92,
|
| 289 |
+
"eval_loss": 0.547610342502594,
|
| 290 |
+
"eval_runtime": 1.7229,
|
| 291 |
+
"eval_samples_per_second": 416.168,
|
| 292 |
+
"eval_steps_per_second": 3.483,
|
| 293 |
+
"eval_wer": 0.2502502502502503,
|
| 294 |
+
"step": 1900
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 18.87,
|
| 298 |
+
"learning_rate": 1.9553367707856575e-05,
|
| 299 |
+
"loss": 0.4257,
|
| 300 |
+
"step": 2000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 18.87,
|
| 304 |
+
"eval_loss": 0.5397851467132568,
|
| 305 |
+
"eval_runtime": 1.7936,
|
| 306 |
+
"eval_samples_per_second": 399.759,
|
| 307 |
+
"eval_steps_per_second": 3.345,
|
| 308 |
+
"eval_wer": 0.2284784784784785,
|
| 309 |
+
"step": 2000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 19.81,
|
| 313 |
+
"learning_rate": 1.9450889074239728e-05,
|
| 314 |
+
"loss": 0.4124,
|
| 315 |
+
"step": 2100
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 19.81,
|
| 319 |
+
"eval_loss": 0.5378283262252808,
|
| 320 |
+
"eval_runtime": 1.6565,
|
| 321 |
+
"eval_samples_per_second": 432.846,
|
| 322 |
+
"eval_steps_per_second": 3.622,
|
| 323 |
+
"eval_wer": 0.2763596930263597,
|
| 324 |
+
"step": 2100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 20.75,
|
| 328 |
+
"learning_rate": 1.9338162504386778e-05,
|
| 329 |
+
"loss": 0.3595,
|
| 330 |
+
"step": 2200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 20.75,
|
| 334 |
+
"eval_loss": 0.5108717679977417,
|
| 335 |
+
"eval_runtime": 1.6671,
|
| 336 |
+
"eval_samples_per_second": 430.078,
|
| 337 |
+
"eval_steps_per_second": 3.599,
|
| 338 |
+
"eval_wer": 0.2147147147147147,
|
| 339 |
+
"step": 2200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 21.7,
|
| 343 |
+
"learning_rate": 1.9215310231739557e-05,
|
| 344 |
+
"loss": 0.3958,
|
| 345 |
+
"step": 2300
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 21.7,
|
| 349 |
+
"eval_loss": 0.4824822247028351,
|
| 350 |
+
"eval_runtime": 1.6672,
|
| 351 |
+
"eval_samples_per_second": 430.059,
|
| 352 |
+
"eval_steps_per_second": 3.599,
|
| 353 |
+
"eval_wer": 0.23423423423423423,
|
| 354 |
+
"step": 2300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 22.64,
|
| 358 |
+
"learning_rate": 1.908246546940074e-05,
|
| 359 |
+
"loss": 0.3546,
|
| 360 |
+
"step": 2400
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 22.64,
|
| 364 |
+
"eval_loss": 0.46492278575897217,
|
| 365 |
+
"eval_runtime": 1.7896,
|
| 366 |
+
"eval_samples_per_second": 400.656,
|
| 367 |
+
"eval_steps_per_second": 3.353,
|
| 368 |
+
"eval_wer": 0.2250583917250584,
|
| 369 |
+
"step": 2400
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 23.58,
|
| 373 |
+
"learning_rate": 1.8939772265686102e-05,
|
| 374 |
+
"loss": 0.304,
|
| 375 |
+
"step": 2500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 23.58,
|
| 379 |
+
"eval_loss": 0.47007447481155396,
|
| 380 |
+
"eval_runtime": 1.7987,
|
| 381 |
+
"eval_samples_per_second": 398.616,
|
| 382 |
+
"eval_steps_per_second": 3.336,
|
| 383 |
+
"eval_wer": 0.21146146146146147,
|
| 384 |
+
"step": 2500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.53,
|
| 388 |
+
"learning_rate": 1.8787385347927804e-05,
|
| 389 |
+
"loss": 0.291,
|
| 390 |
+
"step": 2600
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 24.53,
|
| 394 |
+
"eval_loss": 0.45151692628860474,
|
| 395 |
+
"eval_runtime": 1.6672,
|
| 396 |
+
"eval_samples_per_second": 430.07,
|
| 397 |
+
"eval_steps_per_second": 3.599,
|
| 398 |
+
"eval_wer": 0.21796796796796797,
|
| 399 |
+
"step": 2600
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 25.47,
|
| 403 |
+
"learning_rate": 1.862546995469802e-05,
|
| 404 |
+
"loss": 0.2946,
|
| 405 |
+
"step": 2700
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 25.47,
|
| 409 |
+
"eval_loss": 0.4536628723144531,
|
| 410 |
+
"eval_runtime": 1.6788,
|
| 411 |
+
"eval_samples_per_second": 427.101,
|
| 412 |
+
"eval_steps_per_second": 3.574,
|
| 413 |
+
"eval_wer": 0.2012012012012012,
|
| 414 |
+
"step": 2700
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 26.42,
|
| 418 |
+
"learning_rate": 1.8454201656634855e-05,
|
| 419 |
+
"loss": 0.2588,
|
| 420 |
+
"step": 2800
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 26.42,
|
| 424 |
+
"eval_loss": 0.4422895014286041,
|
| 425 |
+
"eval_runtime": 1.7609,
|
| 426 |
+
"eval_samples_per_second": 407.179,
|
| 427 |
+
"eval_steps_per_second": 3.407,
|
| 428 |
+
"eval_wer": 0.19394394394394393,
|
| 429 |
+
"step": 2800
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 27.36,
|
| 433 |
+
"learning_rate": 1.827376616606485e-05,
|
| 434 |
+
"loss": 0.2625,
|
| 435 |
+
"step": 2900
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 27.36,
|
| 439 |
+
"eval_loss": 0.4493175745010376,
|
| 440 |
+
"eval_runtime": 1.7755,
|
| 441 |
+
"eval_samples_per_second": 403.826,
|
| 442 |
+
"eval_steps_per_second": 3.379,
|
| 443 |
+
"eval_wer": 0.19235902569235902,
|
| 444 |
+
"step": 2900
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 28.3,
|
| 448 |
+
"learning_rate": 1.8084359135628468e-05,
|
| 449 |
+
"loss": 0.2385,
|
| 450 |
+
"step": 3000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 28.3,
|
| 454 |
+
"eval_loss": 0.4364318549633026,
|
| 455 |
+
"eval_runtime": 1.7248,
|
| 456 |
+
"eval_samples_per_second": 415.701,
|
| 457 |
+
"eval_steps_per_second": 3.479,
|
| 458 |
+
"eval_wer": 0.17242242242242242,
|
| 459 |
+
"step": 3000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 29.25,
|
| 463 |
+
"learning_rate": 1.7886185946126956e-05,
|
| 464 |
+
"loss": 0.2327,
|
| 465 |
+
"step": 3100
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 29.25,
|
| 469 |
+
"eval_loss": 0.4382227957248688,
|
| 470 |
+
"eval_runtime": 1.6032,
|
| 471 |
+
"eval_samples_per_second": 447.241,
|
| 472 |
+
"eval_steps_per_second": 3.743,
|
| 473 |
+
"eval_wer": 0.1966966966966967,
|
| 474 |
+
"step": 3100
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 30.19,
|
| 478 |
+
"learning_rate": 1.7679461483820617e-05,
|
| 479 |
+
"loss": 0.26,
|
| 480 |
+
"step": 3200
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 30.19,
|
| 484 |
+
"eval_loss": 0.44537273049354553,
|
| 485 |
+
"eval_runtime": 1.7618,
|
| 486 |
+
"eval_samples_per_second": 406.97,
|
| 487 |
+
"eval_steps_per_second": 3.406,
|
| 488 |
+
"eval_wer": 0.1822655989322656,
|
| 489 |
+
"step": 3200
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 31.13,
|
| 493 |
+
"learning_rate": 1.7464409907419976e-05,
|
| 494 |
+
"loss": 0.2151,
|
| 495 |
+
"step": 3300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 31.13,
|
| 499 |
+
"eval_loss": 0.4423553943634033,
|
| 500 |
+
"eval_runtime": 1.75,
|
| 501 |
+
"eval_samples_per_second": 409.725,
|
| 502 |
+
"eval_steps_per_second": 3.429,
|
| 503 |
+
"eval_wer": 0.1986986986986987,
|
| 504 |
+
"step": 3300
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 32.08,
|
| 508 |
+
"learning_rate": 1.7241264405022477e-05,
|
| 509 |
+
"loss": 0.2213,
|
| 510 |
+
"step": 3400
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 32.08,
|
| 514 |
+
"eval_loss": 0.43766045570373535,
|
| 515 |
+
"eval_runtime": 1.7931,
|
| 516 |
+
"eval_samples_per_second": 399.865,
|
| 517 |
+
"eval_steps_per_second": 3.346,
|
| 518 |
+
"eval_wer": 0.20845845845845845,
|
| 519 |
+
"step": 3400
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 33.02,
|
| 523 |
+
"learning_rate": 1.701026694125833e-05,
|
| 524 |
+
"loss": 0.2226,
|
| 525 |
+
"step": 3500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 33.02,
|
| 529 |
+
"eval_loss": 0.43748462200164795,
|
| 530 |
+
"eval_runtime": 1.7241,
|
| 531 |
+
"eval_samples_per_second": 415.872,
|
| 532 |
+
"eval_steps_per_second": 3.48,
|
| 533 |
+
"eval_wer": 0.20945945945945946,
|
| 534 |
+
"step": 3500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 33.96,
|
| 538 |
+
"learning_rate": 1.6771667994919608e-05,
|
| 539 |
+
"loss": 0.208,
|
| 540 |
+
"step": 3600
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 33.96,
|
| 544 |
+
"eval_loss": 0.43581297993659973,
|
| 545 |
+
"eval_runtime": 1.7973,
|
| 546 |
+
"eval_samples_per_second": 398.933,
|
| 547 |
+
"eval_steps_per_second": 3.338,
|
| 548 |
+
"eval_wer": 0.19944944944944945,
|
| 549 |
+
"step": 3600
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 34.91,
|
| 553 |
+
"learning_rate": 1.6528221181905217e-05,
|
| 554 |
+
"loss": 0.2061,
|
| 555 |
+
"step": 3700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 34.91,
|
| 559 |
+
"eval_loss": 0.43082377314567566,
|
| 560 |
+
"eval_runtime": 1.717,
|
| 561 |
+
"eval_samples_per_second": 417.6,
|
| 562 |
+
"eval_steps_per_second": 3.495,
|
| 563 |
+
"eval_wer": 0.19185852519185853,
|
| 564 |
+
"step": 3700
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 35.85,
|
| 568 |
+
"learning_rate": 1.6275272811746063e-05,
|
| 569 |
+
"loss": 0.1929,
|
| 570 |
+
"step": 3800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 35.85,
|
| 574 |
+
"eval_loss": 0.4298331141471863,
|
| 575 |
+
"eval_runtime": 1.7015,
|
| 576 |
+
"eval_samples_per_second": 421.396,
|
| 577 |
+
"eval_steps_per_second": 3.526,
|
| 578 |
+
"eval_wer": 0.19052385719052387,
|
| 579 |
+
"step": 3800
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 36.79,
|
| 583 |
+
"learning_rate": 1.6015519939371957e-05,
|
| 584 |
+
"loss": 0.1786,
|
| 585 |
+
"step": 3900
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 36.79,
|
| 589 |
+
"eval_loss": 0.4138873815536499,
|
| 590 |
+
"eval_runtime": 1.5956,
|
| 591 |
+
"eval_samples_per_second": 449.369,
|
| 592 |
+
"eval_steps_per_second": 3.76,
|
| 593 |
+
"eval_wer": 0.19744744744744744,
|
| 594 |
+
"step": 3900
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 37.74,
|
| 598 |
+
"learning_rate": 1.5749244224090537e-05,
|
| 599 |
+
"loss": 0.172,
|
| 600 |
+
"step": 4000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 37.74,
|
| 604 |
+
"eval_loss": 0.4183307886123657,
|
| 605 |
+
"eval_runtime": 1.8563,
|
| 606 |
+
"eval_samples_per_second": 386.257,
|
| 607 |
+
"eval_steps_per_second": 3.232,
|
| 608 |
+
"eval_wer": 0.1822655989322656,
|
| 609 |
+
"step": 4000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 38.68,
|
| 613 |
+
"learning_rate": 1.5476734398160752e-05,
|
| 614 |
+
"loss": 0.1769,
|
| 615 |
+
"step": 4100
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 38.68,
|
| 619 |
+
"eval_loss": 0.42518991231918335,
|
| 620 |
+
"eval_runtime": 1.7858,
|
| 621 |
+
"eval_samples_per_second": 401.502,
|
| 622 |
+
"eval_steps_per_second": 3.36,
|
| 623 |
+
"eval_wer": 0.18902235568902234,
|
| 624 |
+
"step": 4100
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 39.62,
|
| 628 |
+
"learning_rate": 1.5198285953710154e-05,
|
| 629 |
+
"loss": 0.1813,
|
| 630 |
+
"step": 4200
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 39.62,
|
| 634 |
+
"eval_loss": 0.4359680712223053,
|
| 635 |
+
"eval_runtime": 1.708,
|
| 636 |
+
"eval_samples_per_second": 419.781,
|
| 637 |
+
"eval_steps_per_second": 3.513,
|
| 638 |
+
"eval_wer": 0.18802135468802136,
|
| 639 |
+
"step": 4200
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 40.57,
|
| 643 |
+
"learning_rate": 1.4914200822322255e-05,
|
| 644 |
+
"loss": 0.1676,
|
| 645 |
+
"step": 4300
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 40.57,
|
| 649 |
+
"eval_loss": 0.4325263500213623,
|
| 650 |
+
"eval_runtime": 1.7629,
|
| 651 |
+
"eval_samples_per_second": 406.719,
|
| 652 |
+
"eval_steps_per_second": 3.404,
|
| 653 |
+
"eval_wer": 0.17701034367701035,
|
| 654 |
+
"step": 4300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 41.51,
|
| 658 |
+
"learning_rate": 1.462478704764133e-05,
|
| 659 |
+
"loss": 0.1581,
|
| 660 |
+
"step": 4400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 41.51,
|
| 664 |
+
"eval_loss": 0.43861034512519836,
|
| 665 |
+
"eval_runtime": 1.6659,
|
| 666 |
+
"eval_samples_per_second": 430.388,
|
| 667 |
+
"eval_steps_per_second": 3.602,
|
| 668 |
+
"eval_wer": 0.17550884217550886,
|
| 669 |
+
"step": 4400
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 42.45,
|
| 673 |
+
"learning_rate": 1.4330358451349708e-05,
|
| 674 |
+
"loss": 0.17,
|
| 675 |
+
"step": 4500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 42.45,
|
| 679 |
+
"eval_loss": 0.4374006986618042,
|
| 680 |
+
"eval_runtime": 1.6606,
|
| 681 |
+
"eval_samples_per_second": 431.774,
|
| 682 |
+
"eval_steps_per_second": 3.613,
|
| 683 |
+
"eval_wer": 0.19794794794794796,
|
| 684 |
+
"step": 4500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 43.4,
|
| 688 |
+
"learning_rate": 1.4031234292879726e-05,
|
| 689 |
+
"loss": 0.1778,
|
| 690 |
+
"step": 4600
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 43.4,
|
| 694 |
+
"eval_loss": 0.43595173954963684,
|
| 695 |
+
"eval_runtime": 1.8676,
|
| 696 |
+
"eval_samples_per_second": 383.907,
|
| 697 |
+
"eval_steps_per_second": 3.213,
|
| 698 |
+
"eval_wer": 0.17258925592258925,
|
| 699 |
+
"step": 4600
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 44.34,
|
| 703 |
+
"learning_rate": 1.3727738923229319e-05,
|
| 704 |
+
"loss": 0.162,
|
| 705 |
+
"step": 4700
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 44.34,
|
| 709 |
+
"eval_loss": 0.44236594438552856,
|
| 710 |
+
"eval_runtime": 1.7683,
|
| 711 |
+
"eval_samples_per_second": 405.482,
|
| 712 |
+
"eval_steps_per_second": 3.393,
|
| 713 |
+
"eval_wer": 0.18218218218218218,
|
| 714 |
+
"step": 4700
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 45.28,
|
| 718 |
+
"learning_rate": 1.342020143325669e-05,
|
| 719 |
+
"loss": 0.1605,
|
| 720 |
+
"step": 4800
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 45.28,
|
| 724 |
+
"eval_loss": 0.45002317428588867,
|
| 725 |
+
"eval_runtime": 1.7462,
|
| 726 |
+
"eval_samples_per_second": 410.616,
|
| 727 |
+
"eval_steps_per_second": 3.436,
|
| 728 |
+
"eval_wer": 0.20645645645645647,
|
| 729 |
+
"step": 4800
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 46.23,
|
| 733 |
+
"learning_rate": 1.3108955296835313e-05,
|
| 734 |
+
"loss": 0.1472,
|
| 735 |
+
"step": 4900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 46.23,
|
| 739 |
+
"eval_loss": 0.4555417597293854,
|
| 740 |
+
"eval_runtime": 1.7133,
|
| 741 |
+
"eval_samples_per_second": 418.491,
|
| 742 |
+
"eval_steps_per_second": 3.502,
|
| 743 |
+
"eval_wer": 0.21021021021021022,
|
| 744 |
+
"step": 4900
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 47.17,
|
| 748 |
+
"learning_rate": 1.2794338009256353e-05,
|
| 749 |
+
"loss": 0.1428,
|
| 750 |
+
"step": 5000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 47.17,
|
| 754 |
+
"eval_loss": 0.4357749819755554,
|
| 755 |
+
"eval_runtime": 1.7755,
|
| 756 |
+
"eval_samples_per_second": 403.823,
|
| 757 |
+
"eval_steps_per_second": 3.379,
|
| 758 |
+
"eval_wer": 0.17334000667334,
|
| 759 |
+
"step": 5000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 48.11,
|
| 763 |
+
"learning_rate": 1.2476690721270442e-05,
|
| 764 |
+
"loss": 0.1393,
|
| 765 |
+
"step": 5100
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 48.11,
|
| 769 |
+
"eval_loss": 0.44057783484458923,
|
| 770 |
+
"eval_runtime": 1.7855,
|
| 771 |
+
"eval_samples_per_second": 401.567,
|
| 772 |
+
"eval_steps_per_second": 3.36,
|
| 773 |
+
"eval_wer": 0.19035702369035704,
|
| 774 |
+
"step": 5100
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 49.06,
|
| 778 |
+
"learning_rate": 1.2156357869165771e-05,
|
| 779 |
+
"loss": 0.1444,
|
| 780 |
+
"step": 5200
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 49.06,
|
| 784 |
+
"eval_loss": 0.4481334090232849,
|
| 785 |
+
"eval_runtime": 1.8611,
|
| 786 |
+
"eval_samples_per_second": 385.26,
|
| 787 |
+
"eval_steps_per_second": 3.224,
|
| 788 |
+
"eval_wer": 0.20303636970303637,
|
| 789 |
+
"step": 5200
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 50.0,
|
| 793 |
+
"learning_rate": 1.1833686801283536e-05,
|
| 794 |
+
"loss": 0.1401,
|
| 795 |
+
"step": 5300
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 50.0,
|
| 799 |
+
"eval_loss": 0.4507039785385132,
|
| 800 |
+
"eval_runtime": 1.8047,
|
| 801 |
+
"eval_samples_per_second": 397.291,
|
| 802 |
+
"eval_steps_per_second": 3.325,
|
| 803 |
+
"eval_wer": 0.19519519519519518,
|
| 804 |
+
"step": 5300
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 50.94,
|
| 808 |
+
"learning_rate": 1.1509027401375717e-05,
|
| 809 |
+
"loss": 0.1311,
|
| 810 |
+
"step": 5400
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 50.94,
|
| 814 |
+
"eval_loss": 0.43529531359672546,
|
| 815 |
+
"eval_runtime": 1.841,
|
| 816 |
+
"eval_samples_per_second": 389.468,
|
| 817 |
+
"eval_steps_per_second": 3.259,
|
| 818 |
+
"eval_wer": 0.1856856856856857,
|
| 819 |
+
"step": 5400
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 51.89,
|
| 823 |
+
"learning_rate": 1.1182731709213658e-05,
|
| 824 |
+
"loss": 0.1337,
|
| 825 |
+
"step": 5500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 51.89,
|
| 829 |
+
"eval_loss": 0.44389182329177856,
|
| 830 |
+
"eval_runtime": 1.814,
|
| 831 |
+
"eval_samples_per_second": 395.264,
|
| 832 |
+
"eval_steps_per_second": 3.308,
|
| 833 |
+
"eval_wer": 0.20178511845178512,
|
| 834 |
+
"step": 5500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 52.83,
|
| 838 |
+
"learning_rate": 1.0855153538858787e-05,
|
| 839 |
+
"loss": 0.1289,
|
| 840 |
+
"step": 5600
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 52.83,
|
| 844 |
+
"eval_loss": 0.44612860679626465,
|
| 845 |
+
"eval_runtime": 1.8003,
|
| 846 |
+
"eval_samples_per_second": 398.261,
|
| 847 |
+
"eval_steps_per_second": 3.333,
|
| 848 |
+
"eval_wer": 0.19461127794461128,
|
| 849 |
+
"step": 5600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 53.77,
|
| 853 |
+
"learning_rate": 1.0526648095009385e-05,
|
| 854 |
+
"loss": 0.1234,
|
| 855 |
+
"step": 5700
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 53.77,
|
| 859 |
+
"eval_loss": 0.4394504427909851,
|
| 860 |
+
"eval_runtime": 1.8352,
|
| 861 |
+
"eval_samples_per_second": 390.687,
|
| 862 |
+
"eval_steps_per_second": 3.269,
|
| 863 |
+
"eval_wer": 0.2047881214547881,
|
| 864 |
+
"step": 5700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 54.72,
|
| 868 |
+
"learning_rate": 1.0197571587839466e-05,
|
| 869 |
+
"loss": 0.1301,
|
| 870 |
+
"step": 5800
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 54.72,
|
| 874 |
+
"eval_loss": 0.45898574590682983,
|
| 875 |
+
"eval_runtime": 1.7706,
|
| 876 |
+
"eval_samples_per_second": 404.955,
|
| 877 |
+
"eval_steps_per_second": 3.389,
|
| 878 |
+
"eval_wer": 0.21137804471137805,
|
| 879 |
+
"step": 5800
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 55.66,
|
| 883 |
+
"learning_rate": 9.868280846747396e-06,
|
| 884 |
+
"loss": 0.1378,
|
| 885 |
+
"step": 5900
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 55.66,
|
| 889 |
+
"eval_loss": 0.45482325553894043,
|
| 890 |
+
"eval_runtime": 1.7006,
|
| 891 |
+
"eval_samples_per_second": 421.624,
|
| 892 |
+
"eval_steps_per_second": 3.528,
|
| 893 |
+
"eval_wer": 0.21438104771438105,
|
| 894 |
+
"step": 5900
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 56.6,
|
| 898 |
+
"learning_rate": 9.539132933433002e-06,
|
| 899 |
+
"loss": 0.1251,
|
| 900 |
+
"step": 6000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 56.6,
|
| 904 |
+
"eval_loss": 0.4476557672023773,
|
| 905 |
+
"eval_runtime": 1.7851,
|
| 906 |
+
"eval_samples_per_second": 401.663,
|
| 907 |
+
"eval_steps_per_second": 3.361,
|
| 908 |
+
"eval_wer": 0.18768768768768768,
|
| 909 |
+
"step": 6000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 57.55,
|
| 913 |
+
"learning_rate": 9.213767592058406e-06,
|
| 914 |
+
"loss": 0.1224,
|
| 915 |
+
"step": 6100
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 57.55,
|
| 919 |
+
"eval_loss": 0.4477730989456177,
|
| 920 |
+
"eval_runtime": 1.9021,
|
| 921 |
+
"eval_samples_per_second": 376.95,
|
| 922 |
+
"eval_steps_per_second": 3.154,
|
| 923 |
+
"eval_wer": 0.1932766099432766,
|
| 924 |
+
"step": 6100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 58.49,
|
| 928 |
+
"learning_rate": 8.885965190625207e-06,
|
| 929 |
+
"loss": 0.1233,
|
| 930 |
+
"step": 6200
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 58.49,
|
| 934 |
+
"eval_loss": 0.44666004180908203,
|
| 935 |
+
"eval_runtime": 1.7327,
|
| 936 |
+
"eval_samples_per_second": 413.795,
|
| 937 |
+
"eval_steps_per_second": 3.463,
|
| 938 |
+
"eval_wer": 0.18410076743410075,
|
| 939 |
+
"step": 6200
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 59.43,
|
| 943 |
+
"learning_rate": 8.559370776885191e-06,
|
| 944 |
+
"loss": 0.1237,
|
| 945 |
+
"step": 6300
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 59.43,
|
| 949 |
+
"eval_loss": 0.43989476561546326,
|
| 950 |
+
"eval_runtime": 1.7589,
|
| 951 |
+
"eval_samples_per_second": 407.647,
|
| 952 |
+
"eval_steps_per_second": 3.411,
|
| 953 |
+
"eval_wer": 0.18343343343343343,
|
| 954 |
+
"step": 6300
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 60.38,
|
| 958 |
+
"learning_rate": 8.234338488814236e-06,
|
| 959 |
+
"loss": 0.1176,
|
| 960 |
+
"step": 6400
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 60.38,
|
| 964 |
+
"eval_loss": 0.44714105129241943,
|
| 965 |
+
"eval_runtime": 1.7645,
|
| 966 |
+
"eval_samples_per_second": 406.356,
|
| 967 |
+
"eval_steps_per_second": 3.4,
|
| 968 |
+
"eval_wer": 0.2097097097097097,
|
| 969 |
+
"step": 6400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 61.32,
|
| 973 |
+
"learning_rate": 7.91122077051963e-06,
|
| 974 |
+
"loss": 0.1117,
|
| 975 |
+
"step": 6500
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 61.32,
|
| 979 |
+
"eval_loss": 0.458738774061203,
|
| 980 |
+
"eval_runtime": 2.0653,
|
| 981 |
+
"eval_samples_per_second": 347.168,
|
| 982 |
+
"eval_steps_per_second": 2.905,
|
| 983 |
+
"eval_wer": 0.19703036369703036,
|
| 984 |
+
"step": 6500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 62.26,
|
| 988 |
+
"learning_rate": 7.59036799007234e-06,
|
| 989 |
+
"loss": 0.111,
|
| 990 |
+
"step": 6600
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 62.26,
|
| 994 |
+
"eval_loss": 0.4706852436065674,
|
| 995 |
+
"eval_runtime": 1.9076,
|
| 996 |
+
"eval_samples_per_second": 375.871,
|
| 997 |
+
"eval_steps_per_second": 3.145,
|
| 998 |
+
"eval_wer": 0.21021021021021022,
|
| 999 |
+
"step": 6600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 63.21,
|
| 1003 |
+
"learning_rate": 7.272128059590406e-06,
|
| 1004 |
+
"loss": 0.1239,
|
| 1005 |
+
"step": 6700
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 63.21,
|
| 1009 |
+
"eval_loss": 0.45181193947792053,
|
| 1010 |
+
"eval_runtime": 1.7149,
|
| 1011 |
+
"eval_samples_per_second": 418.099,
|
| 1012 |
+
"eval_steps_per_second": 3.499,
|
| 1013 |
+
"eval_wer": 0.1922756089422756,
|
| 1014 |
+
"step": 6700
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 64.15,
|
| 1018 |
+
"learning_rate": 6.956846057985423e-06,
|
| 1019 |
+
"loss": 0.1152,
|
| 1020 |
+
"step": 6800
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 64.15,
|
| 1024 |
+
"eval_loss": 0.4502814710140228,
|
| 1025 |
+
"eval_runtime": 1.6787,
|
| 1026 |
+
"eval_samples_per_second": 427.117,
|
| 1027 |
+
"eval_steps_per_second": 3.574,
|
| 1028 |
+
"eval_wer": 0.1966966966966967,
|
| 1029 |
+
"step": 6800
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 65.09,
|
| 1033 |
+
"learning_rate": 6.647966230845655e-06,
|
| 1034 |
+
"loss": 0.1121,
|
| 1035 |
+
"step": 6900
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 65.09,
|
| 1039 |
+
"eval_loss": 0.4467299282550812,
|
| 1040 |
+
"eval_runtime": 1.9525,
|
| 1041 |
+
"eval_samples_per_second": 367.225,
|
| 1042 |
+
"eval_steps_per_second": 3.073,
|
| 1043 |
+
"eval_wer": 0.19444444444444445,
|
| 1044 |
+
"step": 6900
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 66.04,
|
| 1048 |
+
"learning_rate": 6.339584080634824e-06,
|
| 1049 |
+
"loss": 0.1175,
|
| 1050 |
+
"step": 7000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 66.04,
|
| 1054 |
+
"eval_loss": 0.44862234592437744,
|
| 1055 |
+
"eval_runtime": 1.8062,
|
| 1056 |
+
"eval_samples_per_second": 396.962,
|
| 1057 |
+
"eval_steps_per_second": 3.322,
|
| 1058 |
+
"eval_wer": 0.19135802469135801,
|
| 1059 |
+
"step": 7000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 66.98,
|
| 1063 |
+
"learning_rate": 6.03517105001101e-06,
|
| 1064 |
+
"loss": 0.1242,
|
| 1065 |
+
"step": 7100
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 66.98,
|
| 1069 |
+
"eval_loss": 0.4537147581577301,
|
| 1070 |
+
"eval_runtime": 1.7725,
|
| 1071 |
+
"eval_samples_per_second": 404.517,
|
| 1072 |
+
"eval_steps_per_second": 3.385,
|
| 1073 |
+
"eval_wer": 0.1972806139472806,
|
| 1074 |
+
"step": 7100
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 67.92,
|
| 1078 |
+
"learning_rate": 5.735057224884939e-06,
|
| 1079 |
+
"loss": 0.111,
|
| 1080 |
+
"step": 7200
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 67.92,
|
| 1084 |
+
"eval_loss": 0.45872873067855835,
|
| 1085 |
+
"eval_runtime": 1.7013,
|
| 1086 |
+
"eval_samples_per_second": 421.444,
|
| 1087 |
+
"eval_steps_per_second": 3.527,
|
| 1088 |
+
"eval_wer": 0.20078411745078412,
|
| 1089 |
+
"step": 7200
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 68.87,
|
| 1093 |
+
"learning_rate": 5.439568029385422e-06,
|
| 1094 |
+
"loss": 0.1063,
|
| 1095 |
+
"step": 7300
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 68.87,
|
| 1099 |
+
"eval_loss": 0.4550780951976776,
|
| 1100 |
+
"eval_runtime": 1.6746,
|
| 1101 |
+
"eval_samples_per_second": 428.162,
|
| 1102 |
+
"eval_steps_per_second": 3.583,
|
| 1103 |
+
"eval_wer": 0.19285952619285954,
|
| 1104 |
+
"step": 7300
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 69.81,
|
| 1108 |
+
"learning_rate": 5.14902387299036e-06,
|
| 1109 |
+
"loss": 0.1133,
|
| 1110 |
+
"step": 7400
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 69.81,
|
| 1114 |
+
"eval_loss": 0.4546915590763092,
|
| 1115 |
+
"eval_runtime": 1.7468,
|
| 1116 |
+
"eval_samples_per_second": 410.464,
|
| 1117 |
+
"eval_steps_per_second": 3.435,
|
| 1118 |
+
"eval_wer": 0.19285952619285954,
|
| 1119 |
+
"step": 7400
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 70.75,
|
| 1123 |
+
"learning_rate": 4.863739803095299e-06,
|
| 1124 |
+
"loss": 0.1098,
|
| 1125 |
+
"step": 7500
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 70.75,
|
| 1129 |
+
"eval_loss": 0.4511526823043823,
|
| 1130 |
+
"eval_runtime": 1.8195,
|
| 1131 |
+
"eval_samples_per_second": 394.06,
|
| 1132 |
+
"eval_steps_per_second": 3.298,
|
| 1133 |
+
"eval_wer": 0.1981981981981982,
|
| 1134 |
+
"step": 7500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 71.7,
|
| 1138 |
+
"learning_rate": 4.584025163396323e-06,
|
| 1139 |
+
"loss": 0.1123,
|
| 1140 |
+
"step": 7600
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 71.7,
|
| 1144 |
+
"eval_loss": 0.45775842666625977,
|
| 1145 |
+
"eval_runtime": 1.7257,
|
| 1146 |
+
"eval_samples_per_second": 415.482,
|
| 1147 |
+
"eval_steps_per_second": 3.477,
|
| 1148 |
+
"eval_wer": 0.19552886219552887,
|
| 1149 |
+
"step": 7600
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 72.64,
|
| 1153 |
+
"learning_rate": 4.310183258457632e-06,
|
| 1154 |
+
"loss": 0.1144,
|
| 1155 |
+
"step": 7700
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 72.64,
|
| 1159 |
+
"eval_loss": 0.4533255994319916,
|
| 1160 |
+
"eval_runtime": 1.8276,
|
| 1161 |
+
"eval_samples_per_second": 392.326,
|
| 1162 |
+
"eval_steps_per_second": 3.283,
|
| 1163 |
+
"eval_wer": 0.18301634968301636,
|
| 1164 |
+
"step": 7700
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 73.58,
|
| 1168 |
+
"learning_rate": 4.042511024827573e-06,
|
| 1169 |
+
"loss": 0.1113,
|
| 1170 |
+
"step": 7800
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 73.58,
|
| 1174 |
+
"eval_loss": 0.45449498295783997,
|
| 1175 |
+
"eval_runtime": 1.8744,
|
| 1176 |
+
"eval_samples_per_second": 382.518,
|
| 1177 |
+
"eval_steps_per_second": 3.201,
|
| 1178 |
+
"eval_wer": 0.1788455121788455,
|
| 1179 |
+
"step": 7800
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 74.53,
|
| 1183 |
+
"learning_rate": 3.7812987090597696e-06,
|
| 1184 |
+
"loss": 0.0968,
|
| 1185 |
+
"step": 7900
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 74.53,
|
| 1189 |
+
"eval_loss": 0.4584340453147888,
|
| 1190 |
+
"eval_runtime": 1.9007,
|
| 1191 |
+
"eval_samples_per_second": 377.222,
|
| 1192 |
+
"eval_steps_per_second": 3.157,
|
| 1193 |
+
"eval_wer": 0.17250583917250584,
|
| 1194 |
+
"step": 7900
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 75.47,
|
| 1198 |
+
"learning_rate": 3.5268295529884077e-06,
|
| 1199 |
+
"loss": 0.0951,
|
| 1200 |
+
"step": 8000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 75.47,
|
| 1204 |
+
"eval_loss": 0.4645984172821045,
|
| 1205 |
+
"eval_runtime": 1.7512,
|
| 1206 |
+
"eval_samples_per_second": 409.43,
|
| 1207 |
+
"eval_steps_per_second": 3.426,
|
| 1208 |
+
"eval_wer": 0.18593593593593594,
|
| 1209 |
+
"step": 8000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 76.42,
|
| 1213 |
+
"learning_rate": 3.2793794865990092e-06,
|
| 1214 |
+
"loss": 0.0982,
|
| 1215 |
+
"step": 8100
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 76.42,
|
| 1219 |
+
"eval_loss": 0.4557442367076874,
|
| 1220 |
+
"eval_runtime": 1.7832,
|
| 1221 |
+
"eval_samples_per_second": 402.096,
|
| 1222 |
+
"eval_steps_per_second": 3.365,
|
| 1223 |
+
"eval_wer": 0.1812645979312646,
|
| 1224 |
+
"step": 8100
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 77.36,
|
| 1228 |
+
"learning_rate": 3.039216828827717e-06,
|
| 1229 |
+
"loss": 0.0959,
|
| 1230 |
+
"step": 8200
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 77.36,
|
| 1234 |
+
"eval_loss": 0.4565584659576416,
|
| 1235 |
+
"eval_runtime": 1.9633,
|
| 1236 |
+
"eval_samples_per_second": 365.201,
|
| 1237 |
+
"eval_steps_per_second": 3.056,
|
| 1238 |
+
"eval_wer": 0.17417417417417416,
|
| 1239 |
+
"step": 8200
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 78.3,
|
| 1243 |
+
"learning_rate": 2.8066019966134907e-06,
|
| 1244 |
+
"loss": 0.093,
|
| 1245 |
+
"step": 8300
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 78.3,
|
| 1249 |
+
"eval_loss": 0.46041345596313477,
|
| 1250 |
+
"eval_runtime": 1.8183,
|
| 1251 |
+
"eval_samples_per_second": 394.333,
|
| 1252 |
+
"eval_steps_per_second": 3.3,
|
| 1253 |
+
"eval_wer": 0.18802135468802136,
|
| 1254 |
+
"step": 8300
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 79.25,
|
| 1258 |
+
"learning_rate": 2.58178722251872e-06,
|
| 1259 |
+
"loss": 0.103,
|
| 1260 |
+
"step": 8400
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 79.25,
|
| 1264 |
+
"eval_loss": 0.4613533020019531,
|
| 1265 |
+
"eval_runtime": 1.7006,
|
| 1266 |
+
"eval_samples_per_second": 421.607,
|
| 1267 |
+
"eval_steps_per_second": 3.528,
|
| 1268 |
+
"eval_wer": 0.1907741074407741,
|
| 1269 |
+
"step": 8400
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 80.19,
|
| 1273 |
+
"learning_rate": 2.3650162812244725e-06,
|
| 1274 |
+
"loss": 0.1101,
|
| 1275 |
+
"step": 8500
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 80.19,
|
| 1279 |
+
"eval_loss": 0.45864978432655334,
|
| 1280 |
+
"eval_runtime": 1.8618,
|
| 1281 |
+
"eval_samples_per_second": 385.101,
|
| 1282 |
+
"eval_steps_per_second": 3.223,
|
| 1283 |
+
"eval_wer": 0.18051384718051386,
|
| 1284 |
+
"step": 8500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 81.13,
|
| 1288 |
+
"learning_rate": 2.1565242251969022e-06,
|
| 1289 |
+
"loss": 0.1046,
|
| 1290 |
+
"step": 8600
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 81.13,
|
| 1294 |
+
"eval_loss": 0.4590305984020233,
|
| 1295 |
+
"eval_runtime": 1.7322,
|
| 1296 |
+
"eval_samples_per_second": 413.924,
|
| 1297 |
+
"eval_steps_per_second": 3.464,
|
| 1298 |
+
"eval_wer": 0.18251584918251584,
|
| 1299 |
+
"step": 8600
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 82.08,
|
| 1303 |
+
"learning_rate": 1.9565371298114666e-06,
|
| 1304 |
+
"loss": 0.0979,
|
| 1305 |
+
"step": 8700
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 82.08,
|
| 1309 |
+
"eval_loss": 0.4555116593837738,
|
| 1310 |
+
"eval_runtime": 1.9155,
|
| 1311 |
+
"eval_samples_per_second": 374.321,
|
| 1312 |
+
"eval_steps_per_second": 3.132,
|
| 1313 |
+
"eval_wer": 0.17617617617617617,
|
| 1314 |
+
"step": 8700
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 83.02,
|
| 1318 |
+
"learning_rate": 1.7652718482113417e-06,
|
| 1319 |
+
"loss": 0.103,
|
| 1320 |
+
"step": 8800
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 83.02,
|
| 1324 |
+
"eval_loss": 0.4572655260562897,
|
| 1325 |
+
"eval_runtime": 1.7332,
|
| 1326 |
+
"eval_samples_per_second": 413.676,
|
| 1327 |
+
"eval_steps_per_second": 3.462,
|
| 1328 |
+
"eval_wer": 0.17801134467801136,
|
| 1329 |
+
"step": 8800
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 83.96,
|
| 1333 |
+
"learning_rate": 1.5829357761658214e-06,
|
| 1334 |
+
"loss": 0.0958,
|
| 1335 |
+
"step": 8900
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 83.96,
|
| 1339 |
+
"eval_loss": 0.4574595093727112,
|
| 1340 |
+
"eval_runtime": 1.7971,
|
| 1341 |
+
"eval_samples_per_second": 398.975,
|
| 1342 |
+
"eval_steps_per_second": 3.339,
|
| 1343 |
+
"eval_wer": 0.1802635969302636,
|
| 1344 |
+
"step": 8900
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 84.91,
|
| 1348 |
+
"learning_rate": 1.4097266271836695e-06,
|
| 1349 |
+
"loss": 0.0948,
|
| 1350 |
+
"step": 9000
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 84.91,
|
| 1354 |
+
"eval_loss": 0.45812544226646423,
|
| 1355 |
+
"eval_runtime": 2.0018,
|
| 1356 |
+
"eval_samples_per_second": 358.171,
|
| 1357 |
+
"eval_steps_per_second": 2.997,
|
| 1358 |
+
"eval_wer": 0.18143143143143142,
|
| 1359 |
+
"step": 9000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 85.85,
|
| 1363 |
+
"learning_rate": 1.245832218125328e-06,
|
| 1364 |
+
"loss": 0.1003,
|
| 1365 |
+
"step": 9100
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 85.85,
|
| 1369 |
+
"eval_loss": 0.4599616527557373,
|
| 1370 |
+
"eval_runtime": 1.7462,
|
| 1371 |
+
"eval_samples_per_second": 410.601,
|
| 1372 |
+
"eval_steps_per_second": 3.436,
|
| 1373 |
+
"eval_wer": 0.18301634968301636,
|
| 1374 |
+
"step": 9100
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 86.79,
|
| 1378 |
+
"learning_rate": 1.0914302655463837e-06,
|
| 1379 |
+
"loss": 0.1066,
|
| 1380 |
+
"step": 9200
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 86.79,
|
| 1384 |
+
"eval_loss": 0.46090590953826904,
|
| 1385 |
+
"eval_runtime": 1.7826,
|
| 1386 |
+
"eval_samples_per_second": 402.218,
|
| 1387 |
+
"eval_steps_per_second": 3.366,
|
| 1388 |
+
"eval_wer": 0.18702035368702036,
|
| 1389 |
+
"step": 9200
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 87.74,
|
| 1393 |
+
"learning_rate": 9.466881929931582e-07,
|
| 1394 |
+
"loss": 0.0887,
|
| 1395 |
+
"step": 9300
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 87.74,
|
| 1399 |
+
"eval_loss": 0.4614775776863098,
|
| 1400 |
+
"eval_runtime": 1.7973,
|
| 1401 |
+
"eval_samples_per_second": 398.922,
|
| 1402 |
+
"eval_steps_per_second": 3.338,
|
| 1403 |
+
"eval_wer": 0.18335001668335002,
|
| 1404 |
+
"step": 9300
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 88.68,
|
| 1408 |
+
"learning_rate": 8.117629494594015e-07,
|
| 1409 |
+
"loss": 0.0936,
|
| 1410 |
+
"step": 9400
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 88.68,
|
| 1414 |
+
"eval_loss": 0.4609696865081787,
|
| 1415 |
+
"eval_runtime": 1.8908,
|
| 1416 |
+
"eval_samples_per_second": 379.212,
|
| 1417 |
+
"eval_steps_per_second": 3.173,
|
| 1418 |
+
"eval_wer": 0.18193193193193194,
|
| 1419 |
+
"step": 9400
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 89.62,
|
| 1423 |
+
"learning_rate": 6.86800839200884e-07,
|
| 1424 |
+
"loss": 0.0892,
|
| 1425 |
+
"step": 9500
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 89.62,
|
| 1429 |
+
"eval_loss": 0.4594886898994446,
|
| 1430 |
+
"eval_runtime": 1.7755,
|
| 1431 |
+
"eval_samples_per_second": 403.82,
|
| 1432 |
+
"eval_steps_per_second": 3.379,
|
| 1433 |
+
"eval_wer": 0.18009676343009676,
|
| 1434 |
+
"step": 9500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 90.57,
|
| 1438 |
+
"learning_rate": 5.719373630924741e-07,
|
| 1439 |
+
"loss": 0.1039,
|
| 1440 |
+
"step": 9600
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 90.57,
|
| 1444 |
+
"eval_loss": 0.461166113615036,
|
| 1445 |
+
"eval_runtime": 1.8202,
|
| 1446 |
+
"eval_samples_per_second": 393.913,
|
| 1447 |
+
"eval_steps_per_second": 3.296,
|
| 1448 |
+
"eval_wer": 0.18368368368368368,
|
| 1449 |
+
"step": 9600
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 91.51,
|
| 1453 |
+
"learning_rate": 4.672970716997094e-07,
|
| 1454 |
+
"loss": 0.097,
|
| 1455 |
+
"step": 9700
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 91.51,
|
| 1459 |
+
"eval_loss": 0.4609868824481964,
|
| 1460 |
+
"eval_runtime": 1.8574,
|
| 1461 |
+
"eval_samples_per_second": 386.029,
|
| 1462 |
+
"eval_steps_per_second": 3.23,
|
| 1463 |
+
"eval_wer": 0.18343343343343343,
|
| 1464 |
+
"step": 9700
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 92.45,
|
| 1468 |
+
"learning_rate": 3.73884955111915e-07,
|
| 1469 |
+
"loss": 0.0969,
|
| 1470 |
+
"step": 9800
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 92.45,
|
| 1474 |
+
"eval_loss": 0.4604853093624115,
|
| 1475 |
+
"eval_runtime": 1.8926,
|
| 1476 |
+
"eval_samples_per_second": 378.848,
|
| 1477 |
+
"eval_steps_per_second": 3.17,
|
| 1478 |
+
"eval_wer": 0.18435101768435103,
|
| 1479 |
+
"step": 9800
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 93.4,
|
| 1483 |
+
"learning_rate": 2.8991536219946546e-07,
|
| 1484 |
+
"loss": 0.0946,
|
| 1485 |
+
"step": 9900
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 93.4,
|
| 1489 |
+
"eval_loss": 0.4596187174320221,
|
| 1490 |
+
"eval_runtime": 1.7445,
|
| 1491 |
+
"eval_samples_per_second": 411.004,
|
| 1492 |
+
"eval_steps_per_second": 3.439,
|
| 1493 |
+
"eval_wer": 0.1842676009342676,
|
| 1494 |
+
"step": 9900
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 94.34,
|
| 1498 |
+
"learning_rate": 2.164747605209294e-07,
|
| 1499 |
+
"loss": 0.0947,
|
| 1500 |
+
"step": 10000
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 94.34,
|
| 1504 |
+
"eval_loss": 0.4604862630367279,
|
| 1505 |
+
"eval_runtime": 1.8793,
|
| 1506 |
+
"eval_samples_per_second": 381.515,
|
| 1507 |
+
"eval_steps_per_second": 3.193,
|
| 1508 |
+
"eval_wer": 0.18501835168501835,
|
| 1509 |
+
"step": 10000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 95.28,
|
| 1513 |
+
"learning_rate": 1.5364278434110592e-07,
|
| 1514 |
+
"loss": 0.095,
|
| 1515 |
+
"step": 10100
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 95.28,
|
| 1519 |
+
"eval_loss": 0.46158939599990845,
|
| 1520 |
+
"eval_runtime": 1.9246,
|
| 1521 |
+
"eval_samples_per_second": 372.547,
|
| 1522 |
+
"eval_steps_per_second": 3.118,
|
| 1523 |
+
"eval_wer": 0.18610276943610277,
|
| 1524 |
+
"step": 10100
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 96.23,
|
| 1528 |
+
"learning_rate": 1.0148756461364706e-07,
|
| 1529 |
+
"loss": 0.0856,
|
| 1530 |
+
"step": 10200
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 96.23,
|
| 1534 |
+
"eval_loss": 0.4610789716243744,
|
| 1535 |
+
"eval_runtime": 1.7548,
|
| 1536 |
+
"eval_samples_per_second": 408.584,
|
| 1537 |
+
"eval_steps_per_second": 3.419,
|
| 1538 |
+
"eval_wer": 0.1852686019352686,
|
| 1539 |
+
"step": 10200
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 97.17,
|
| 1543 |
+
"learning_rate": 6.006565510423112e-08,
|
| 1544 |
+
"loss": 0.0983,
|
| 1545 |
+
"step": 10300
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 97.17,
|
| 1549 |
+
"eval_loss": 0.4602554440498352,
|
| 1550 |
+
"eval_runtime": 1.7944,
|
| 1551 |
+
"eval_samples_per_second": 399.587,
|
| 1552 |
+
"eval_steps_per_second": 3.344,
|
| 1553 |
+
"eval_wer": 0.18501835168501835,
|
| 1554 |
+
"step": 10300
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 98.11,
|
| 1558 |
+
"learning_rate": 2.942197106729383e-08,
|
| 1559 |
+
"loss": 0.0947,
|
| 1560 |
+
"step": 10400
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 98.11,
|
| 1564 |
+
"eval_loss": 0.460513710975647,
|
| 1565 |
+
"eval_runtime": 1.9261,
|
| 1566 |
+
"eval_samples_per_second": 372.264,
|
| 1567 |
+
"eval_steps_per_second": 3.115,
|
| 1568 |
+
"eval_wer": 0.1852686019352686,
|
| 1569 |
+
"step": 10400
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 99.06,
|
| 1573 |
+
"learning_rate": 9.589740542820247e-09,
|
| 1574 |
+
"loss": 0.0948,
|
| 1575 |
+
"step": 10500
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 99.06,
|
| 1579 |
+
"eval_loss": 0.46040719747543335,
|
| 1580 |
+
"eval_runtime": 1.8213,
|
| 1581 |
+
"eval_samples_per_second": 393.677,
|
| 1582 |
+
"eval_steps_per_second": 3.294,
|
| 1583 |
+
"eval_wer": 0.1852686019352686,
|
| 1584 |
+
"step": 10500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 100.0,
|
| 1588 |
+
"learning_rate": 5.904683259938537e-10,
|
| 1589 |
+
"loss": 0.0917,
|
| 1590 |
+
"step": 10600
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 100.0,
|
| 1594 |
+
"eval_loss": 0.4603355824947357,
|
| 1595 |
+
"eval_runtime": 2.1628,
|
| 1596 |
+
"eval_samples_per_second": 331.521,
|
| 1597 |
+
"eval_steps_per_second": 2.774,
|
| 1598 |
+
"eval_wer": 0.18493493493493493,
|
| 1599 |
+
"step": 10600
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 100.0,
|
| 1603 |
+
"step": 10600,
|
| 1604 |
+
"total_flos": 2.842046364754798e+19,
|
| 1605 |
+
"train_loss": 2.246708391477477,
|
| 1606 |
+
"train_runtime": 3748.9999,
|
| 1607 |
+
"train_samples_per_second": 90.264,
|
| 1608 |
+
"train_steps_per_second": 2.827
|
| 1609 |
+
}
|
| 1610 |
+
],
|
| 1611 |
+
"logging_steps": 100,
|
| 1612 |
+
"max_steps": 10600,
|
| 1613 |
+
"num_train_epochs": 100,
|
| 1614 |
+
"save_steps": 100,
|
| 1615 |
+
"total_flos": 2.842046364754798e+19,
|
| 1616 |
+
"trial_name": null,
|
| 1617 |
+
"trial_params": null
|
| 1618 |
+
}
|