End of training
Browse files- trainer_state.json +1618 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1618 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08583583583583583,
|
| 3 |
+
"best_model_checkpoint": "w2v2_ablation_focal_ctc_a0.75_g2.0-best_on-ling_head-tp0.025_tl10_fp0.001_fl16/checkpoint-4400",
|
| 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": 1.2830188679245284e-06,
|
| 14 |
+
"loss": 1337.3802,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.94,
|
| 19 |
+
"eval_loss": 875.6535034179688,
|
| 20 |
+
"eval_runtime": 3.072,
|
| 21 |
+
"eval_samples_per_second": 233.397,
|
| 22 |
+
"eval_steps_per_second": 3.906,
|
| 23 |
+
"eval_wer": 18.64039039039039,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.89,
|
| 28 |
+
"learning_rate": 3.169811320754717e-06,
|
| 29 |
+
"loss": 928.4498,
|
| 30 |
+
"step": 200
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.89,
|
| 34 |
+
"eval_loss": 336.85919189453125,
|
| 35 |
+
"eval_runtime": 2.5032,
|
| 36 |
+
"eval_samples_per_second": 286.428,
|
| 37 |
+
"eval_steps_per_second": 4.794,
|
| 38 |
+
"eval_wer": 17.08541875208542,
|
| 39 |
+
"step": 200
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.83,
|
| 43 |
+
"learning_rate": 5.056603773584906e-06,
|
| 44 |
+
"loss": 159.8141,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.83,
|
| 49 |
+
"eval_loss": 65.91426086425781,
|
| 50 |
+
"eval_runtime": 2.3194,
|
| 51 |
+
"eval_samples_per_second": 309.133,
|
| 52 |
+
"eval_steps_per_second": 5.174,
|
| 53 |
+
"eval_wer": 1.0,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.77,
|
| 58 |
+
"learning_rate": 6.943396226415094e-06,
|
| 59 |
+
"loss": 84.4352,
|
| 60 |
+
"step": 400
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 3.77,
|
| 64 |
+
"eval_loss": 60.37297439575195,
|
| 65 |
+
"eval_runtime": 2.6764,
|
| 66 |
+
"eval_samples_per_second": 267.896,
|
| 67 |
+
"eval_steps_per_second": 4.484,
|
| 68 |
+
"eval_wer": 1.0,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 4.72,
|
| 73 |
+
"learning_rate": 8.830188679245285e-06,
|
| 74 |
+
"loss": 77.6086,
|
| 75 |
+
"step": 500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 4.72,
|
| 79 |
+
"eval_loss": 57.35933303833008,
|
| 80 |
+
"eval_runtime": 2.6396,
|
| 81 |
+
"eval_samples_per_second": 271.633,
|
| 82 |
+
"eval_steps_per_second": 4.546,
|
| 83 |
+
"eval_wer": 1.0,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.66,
|
| 88 |
+
"learning_rate": 1.0716981132075473e-05,
|
| 89 |
+
"loss": 74.6091,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 5.66,
|
| 94 |
+
"eval_loss": 56.16162109375,
|
| 95 |
+
"eval_runtime": 3.0321,
|
| 96 |
+
"eval_samples_per_second": 236.471,
|
| 97 |
+
"eval_steps_per_second": 3.958,
|
| 98 |
+
"eval_wer": 1.0,
|
| 99 |
+
"step": 600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.6,
|
| 103 |
+
"learning_rate": 1.2603773584905661e-05,
|
| 104 |
+
"loss": 73.5983,
|
| 105 |
+
"step": 700
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.6,
|
| 109 |
+
"eval_loss": 55.27739334106445,
|
| 110 |
+
"eval_runtime": 2.353,
|
| 111 |
+
"eval_samples_per_second": 304.721,
|
| 112 |
+
"eval_steps_per_second": 5.1,
|
| 113 |
+
"eval_wer": 1.0,
|
| 114 |
+
"step": 700
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.55,
|
| 118 |
+
"learning_rate": 1.449056603773585e-05,
|
| 119 |
+
"loss": 72.9967,
|
| 120 |
+
"step": 800
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.55,
|
| 124 |
+
"eval_loss": 54.65113067626953,
|
| 125 |
+
"eval_runtime": 2.238,
|
| 126 |
+
"eval_samples_per_second": 320.375,
|
| 127 |
+
"eval_steps_per_second": 5.362,
|
| 128 |
+
"eval_wer": 1.0,
|
| 129 |
+
"step": 800
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 8.49,
|
| 133 |
+
"learning_rate": 1.637735849056604e-05,
|
| 134 |
+
"loss": 71.2266,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.49,
|
| 139 |
+
"eval_loss": 54.53617477416992,
|
| 140 |
+
"eval_runtime": 2.445,
|
| 141 |
+
"eval_samples_per_second": 293.257,
|
| 142 |
+
"eval_steps_per_second": 4.908,
|
| 143 |
+
"eval_wer": 1.0,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.43,
|
| 148 |
+
"learning_rate": 1.8264150943396227e-05,
|
| 149 |
+
"loss": 69.7741,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.43,
|
| 154 |
+
"eval_loss": 51.871822357177734,
|
| 155 |
+
"eval_runtime": 2.3898,
|
| 156 |
+
"eval_samples_per_second": 300.02,
|
| 157 |
+
"eval_steps_per_second": 5.021,
|
| 158 |
+
"eval_wer": 0.9647981314647981,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 10.38,
|
| 163 |
+
"learning_rate": 1.999996529814242e-05,
|
| 164 |
+
"loss": 58.1878,
|
| 165 |
+
"step": 1100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 10.38,
|
| 169 |
+
"eval_loss": 28.900115966796875,
|
| 170 |
+
"eval_runtime": 2.2862,
|
| 171 |
+
"eval_samples_per_second": 313.622,
|
| 172 |
+
"eval_steps_per_second": 5.249,
|
| 173 |
+
"eval_wer": 0.5654821488154822,
|
| 174 |
+
"step": 1100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 11.32,
|
| 178 |
+
"learning_rate": 1.9993676249406895e-05,
|
| 179 |
+
"loss": 32.9238,
|
| 180 |
+
"step": 1200
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 11.32,
|
| 184 |
+
"eval_loss": 12.709677696228027,
|
| 185 |
+
"eval_runtime": 2.4457,
|
| 186 |
+
"eval_samples_per_second": 293.166,
|
| 187 |
+
"eval_steps_per_second": 4.907,
|
| 188 |
+
"eval_wer": 0.2390724057390724,
|
| 189 |
+
"step": 1200
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 12.26,
|
| 193 |
+
"learning_rate": 1.997655070093429e-05,
|
| 194 |
+
"loss": 21.0735,
|
| 195 |
+
"step": 1300
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.26,
|
| 199 |
+
"eval_loss": 8.588455200195312,
|
| 200 |
+
"eval_runtime": 2.5444,
|
| 201 |
+
"eval_samples_per_second": 281.79,
|
| 202 |
+
"eval_steps_per_second": 4.716,
|
| 203 |
+
"eval_wer": 0.17851184517851185,
|
| 204 |
+
"step": 1300
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 13.21,
|
| 208 |
+
"learning_rate": 1.994860722256786e-05,
|
| 209 |
+
"loss": 15.9281,
|
| 210 |
+
"step": 1400
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.21,
|
| 214 |
+
"eval_loss": 6.895908832550049,
|
| 215 |
+
"eval_runtime": 2.3592,
|
| 216 |
+
"eval_samples_per_second": 303.92,
|
| 217 |
+
"eval_steps_per_second": 5.087,
|
| 218 |
+
"eval_wer": 0.1529029029029029,
|
| 219 |
+
"step": 1400
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 14.15,
|
| 223 |
+
"learning_rate": 1.9909876114418242e-05,
|
| 224 |
+
"loss": 13.7108,
|
| 225 |
+
"step": 1500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 14.15,
|
| 229 |
+
"eval_loss": 5.751368045806885,
|
| 230 |
+
"eval_runtime": 2.4597,
|
| 231 |
+
"eval_samples_per_second": 291.499,
|
| 232 |
+
"eval_steps_per_second": 4.879,
|
| 233 |
+
"eval_wer": 0.13922255588922255,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 15.09,
|
| 238 |
+
"learning_rate": 1.9860399374007944e-05,
|
| 239 |
+
"loss": 11.2293,
|
| 240 |
+
"step": 1600
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.09,
|
| 244 |
+
"eval_loss": 4.973855495452881,
|
| 245 |
+
"eval_runtime": 2.554,
|
| 246 |
+
"eval_samples_per_second": 280.739,
|
| 247 |
+
"eval_steps_per_second": 4.699,
|
| 248 |
+
"eval_wer": 0.12437437437437437,
|
| 249 |
+
"step": 1600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.04,
|
| 253 |
+
"learning_rate": 1.980023065073195e-05,
|
| 254 |
+
"loss": 10.3682,
|
| 255 |
+
"step": 1700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 16.04,
|
| 259 |
+
"eval_loss": 4.508381366729736,
|
| 260 |
+
"eval_runtime": 2.2355,
|
| 261 |
+
"eval_samples_per_second": 320.735,
|
| 262 |
+
"eval_steps_per_second": 5.368,
|
| 263 |
+
"eval_wer": 0.12370704037370704,
|
| 264 |
+
"step": 1700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.98,
|
| 268 |
+
"learning_rate": 1.972943518768377e-05,
|
| 269 |
+
"loss": 9.6654,
|
| 270 |
+
"step": 1800
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.98,
|
| 274 |
+
"eval_loss": 4.370257377624512,
|
| 275 |
+
"eval_runtime": 2.461,
|
| 276 |
+
"eval_samples_per_second": 291.343,
|
| 277 |
+
"eval_steps_per_second": 4.876,
|
| 278 |
+
"eval_wer": 0.12587587587587587,
|
| 279 |
+
"step": 1800
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 17.92,
|
| 283 |
+
"learning_rate": 1.964808975090999e-05,
|
| 284 |
+
"loss": 8.816,
|
| 285 |
+
"step": 1900
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 17.92,
|
| 289 |
+
"eval_loss": 4.127786636352539,
|
| 290 |
+
"eval_runtime": 2.5207,
|
| 291 |
+
"eval_samples_per_second": 284.449,
|
| 292 |
+
"eval_steps_per_second": 4.761,
|
| 293 |
+
"eval_wer": 0.11428094761428095,
|
| 294 |
+
"step": 1900
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 18.87,
|
| 298 |
+
"learning_rate": 1.955628254617017e-05,
|
| 299 |
+
"loss": 8.8608,
|
| 300 |
+
"step": 2000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 18.87,
|
| 304 |
+
"eval_loss": 3.910537004470825,
|
| 305 |
+
"eval_runtime": 2.4695,
|
| 306 |
+
"eval_samples_per_second": 290.343,
|
| 307 |
+
"eval_steps_per_second": 4.859,
|
| 308 |
+
"eval_wer": 0.10744077410744077,
|
| 309 |
+
"step": 2000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 19.81,
|
| 313 |
+
"learning_rate": 1.9454113123292133e-05,
|
| 314 |
+
"loss": 7.8629,
|
| 315 |
+
"step": 2100
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 19.81,
|
| 319 |
+
"eval_loss": 3.9113922119140625,
|
| 320 |
+
"eval_runtime": 2.5572,
|
| 321 |
+
"eval_samples_per_second": 280.382,
|
| 322 |
+
"eval_steps_per_second": 4.693,
|
| 323 |
+
"eval_wer": 0.12370704037370704,
|
| 324 |
+
"step": 2100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 20.75,
|
| 328 |
+
"learning_rate": 1.9341692268226572e-05,
|
| 329 |
+
"loss": 7.8569,
|
| 330 |
+
"step": 2200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 20.75,
|
| 334 |
+
"eval_loss": 3.7353532314300537,
|
| 335 |
+
"eval_runtime": 2.4284,
|
| 336 |
+
"eval_samples_per_second": 295.26,
|
| 337 |
+
"eval_steps_per_second": 4.942,
|
| 338 |
+
"eval_wer": 0.11211211211211211,
|
| 339 |
+
"step": 2200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 21.7,
|
| 343 |
+
"learning_rate": 1.921914188291787e-05,
|
| 344 |
+
"loss": 7.3392,
|
| 345 |
+
"step": 2300
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 21.7,
|
| 349 |
+
"eval_loss": 3.666839599609375,
|
| 350 |
+
"eval_runtime": 2.409,
|
| 351 |
+
"eval_samples_per_second": 297.636,
|
| 352 |
+
"eval_steps_per_second": 4.981,
|
| 353 |
+
"eval_wer": 0.10560560560560561,
|
| 354 |
+
"step": 2300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 22.64,
|
| 358 |
+
"learning_rate": 1.908659485312148e-05,
|
| 359 |
+
"loss": 7.2164,
|
| 360 |
+
"step": 2400
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 22.64,
|
| 364 |
+
"eval_loss": 3.5747179985046387,
|
| 365 |
+
"eval_runtime": 2.3787,
|
| 366 |
+
"eval_samples_per_second": 301.422,
|
| 367 |
+
"eval_steps_per_second": 5.045,
|
| 368 |
+
"eval_wer": 0.11277944611277944,
|
| 369 |
+
"step": 2400
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 23.58,
|
| 373 |
+
"learning_rate": 1.894419490431116e-05,
|
| 374 |
+
"loss": 7.2758,
|
| 375 |
+
"step": 2500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 23.58,
|
| 379 |
+
"eval_loss": 3.4933085441589355,
|
| 380 |
+
"eval_runtime": 2.546,
|
| 381 |
+
"eval_samples_per_second": 281.622,
|
| 382 |
+
"eval_steps_per_second": 4.713,
|
| 383 |
+
"eval_wer": 0.1016016016016016,
|
| 384 |
+
"step": 2500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.53,
|
| 388 |
+
"learning_rate": 1.8792096445832317e-05,
|
| 389 |
+
"loss": 6.4516,
|
| 390 |
+
"step": 2600
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 24.53,
|
| 394 |
+
"eval_loss": 3.482123613357544,
|
| 395 |
+
"eval_runtime": 2.3838,
|
| 396 |
+
"eval_samples_per_second": 300.782,
|
| 397 |
+
"eval_steps_per_second": 5.034,
|
| 398 |
+
"eval_wer": 0.09876543209876543,
|
| 399 |
+
"step": 2600
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 25.47,
|
| 403 |
+
"learning_rate": 1.8630464403470435e-05,
|
| 404 |
+
"loss": 6.45,
|
| 405 |
+
"step": 2700
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 25.47,
|
| 409 |
+
"eval_loss": 3.3719635009765625,
|
| 410 |
+
"eval_runtime": 2.362,
|
| 411 |
+
"eval_samples_per_second": 303.558,
|
| 412 |
+
"eval_steps_per_second": 5.08,
|
| 413 |
+
"eval_wer": 0.0995995995995996,
|
| 414 |
+
"step": 2700
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 26.42,
|
| 418 |
+
"learning_rate": 1.8461229667449597e-05,
|
| 419 |
+
"loss": 6.0068,
|
| 420 |
+
"step": 2800
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 26.42,
|
| 424 |
+
"eval_loss": 3.442462205886841,
|
| 425 |
+
"eval_runtime": 2.4309,
|
| 426 |
+
"eval_samples_per_second": 294.954,
|
| 427 |
+
"eval_steps_per_second": 4.936,
|
| 428 |
+
"eval_wer": 0.10443777110443778,
|
| 429 |
+
"step": 2800
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 27.36,
|
| 433 |
+
"learning_rate": 1.828115717353417e-05,
|
| 434 |
+
"loss": 5.5781,
|
| 435 |
+
"step": 2900
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 27.36,
|
| 439 |
+
"eval_loss": 3.3221378326416016,
|
| 440 |
+
"eval_runtime": 2.3098,
|
| 441 |
+
"eval_samples_per_second": 310.419,
|
| 442 |
+
"eval_steps_per_second": 5.195,
|
| 443 |
+
"eval_wer": 0.10143476810143477,
|
| 444 |
+
"step": 2900
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 28.3,
|
| 448 |
+
"learning_rate": 1.809210512541925e-05,
|
| 449 |
+
"loss": 5.5837,
|
| 450 |
+
"step": 3000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 28.3,
|
| 454 |
+
"eval_loss": 3.4974489212036133,
|
| 455 |
+
"eval_runtime": 2.4817,
|
| 456 |
+
"eval_samples_per_second": 288.914,
|
| 457 |
+
"eval_steps_per_second": 4.835,
|
| 458 |
+
"eval_wer": 0.1041041041041041,
|
| 459 |
+
"step": 3000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 29.25,
|
| 463 |
+
"learning_rate": 1.7894278518986088e-05,
|
| 464 |
+
"loss": 5.7895,
|
| 465 |
+
"step": 3100
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 29.25,
|
| 469 |
+
"eval_loss": 3.35359263420105,
|
| 470 |
+
"eval_runtime": 2.5133,
|
| 471 |
+
"eval_samples_per_second": 285.278,
|
| 472 |
+
"eval_steps_per_second": 4.775,
|
| 473 |
+
"eval_wer": 0.09501167834501167,
|
| 474 |
+
"step": 3100
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 30.19,
|
| 478 |
+
"learning_rate": 1.7687891864682602e-05,
|
| 479 |
+
"loss": 5.6272,
|
| 480 |
+
"step": 3200
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 30.19,
|
| 484 |
+
"eval_loss": 3.2035768032073975,
|
| 485 |
+
"eval_runtime": 2.4143,
|
| 486 |
+
"eval_samples_per_second": 296.982,
|
| 487 |
+
"eval_steps_per_second": 4.97,
|
| 488 |
+
"eval_wer": 0.09601267934601268,
|
| 489 |
+
"step": 3200
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 31.13,
|
| 493 |
+
"learning_rate": 1.7473168954922044e-05,
|
| 494 |
+
"loss": 5.594,
|
| 495 |
+
"step": 3300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 31.13,
|
| 499 |
+
"eval_loss": 3.1747331619262695,
|
| 500 |
+
"eval_runtime": 2.332,
|
| 501 |
+
"eval_samples_per_second": 307.466,
|
| 502 |
+
"eval_steps_per_second": 5.146,
|
| 503 |
+
"eval_wer": 0.09125792459125792,
|
| 504 |
+
"step": 3300
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 32.08,
|
| 508 |
+
"learning_rate": 1.7250342621416897e-05,
|
| 509 |
+
"loss": 4.791,
|
| 510 |
+
"step": 3400
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 32.08,
|
| 514 |
+
"eval_loss": 3.12252140045166,
|
| 515 |
+
"eval_runtime": 2.3979,
|
| 516 |
+
"eval_samples_per_second": 299.007,
|
| 517 |
+
"eval_steps_per_second": 5.004,
|
| 518 |
+
"eval_wer": 0.10377043710377044,
|
| 519 |
+
"step": 3400
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 33.02,
|
| 523 |
+
"learning_rate": 1.7019654482711144e-05,
|
| 524 |
+
"loss": 5.0596,
|
| 525 |
+
"step": 3500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 33.02,
|
| 529 |
+
"eval_loss": 3.2113351821899414,
|
| 530 |
+
"eval_runtime": 2.2358,
|
| 531 |
+
"eval_samples_per_second": 320.694,
|
| 532 |
+
"eval_steps_per_second": 5.367,
|
| 533 |
+
"eval_wer": 0.1095261928595262,
|
| 534 |
+
"step": 3500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 33.96,
|
| 538 |
+
"learning_rate": 1.6781354682184668e-05,
|
| 539 |
+
"loss": 4.985,
|
| 540 |
+
"step": 3600
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 33.96,
|
| 544 |
+
"eval_loss": 3.062237024307251,
|
| 545 |
+
"eval_runtime": 2.3928,
|
| 546 |
+
"eval_samples_per_second": 299.643,
|
| 547 |
+
"eval_steps_per_second": 5.015,
|
| 548 |
+
"eval_wer": 0.09292625959292626,
|
| 549 |
+
"step": 3600
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 34.91,
|
| 553 |
+
"learning_rate": 1.6535701616813845e-05,
|
| 554 |
+
"loss": 4.731,
|
| 555 |
+
"step": 3700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 34.91,
|
| 559 |
+
"eval_loss": 3.0940449237823486,
|
| 560 |
+
"eval_runtime": 2.4517,
|
| 561 |
+
"eval_samples_per_second": 292.45,
|
| 562 |
+
"eval_steps_per_second": 4.895,
|
| 563 |
+
"eval_wer": 0.09559559559559559,
|
| 564 |
+
"step": 3700
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 35.85,
|
| 568 |
+
"learning_rate": 1.628296165698249e-05,
|
| 569 |
+
"loss": 4.6287,
|
| 570 |
+
"step": 3800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 35.85,
|
| 574 |
+
"eval_loss": 3.0453147888183594,
|
| 575 |
+
"eval_runtime": 2.5566,
|
| 576 |
+
"eval_samples_per_second": 280.449,
|
| 577 |
+
"eval_steps_per_second": 4.694,
|
| 578 |
+
"eval_wer": 0.0960960960960961,
|
| 579 |
+
"step": 3800
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 36.79,
|
| 583 |
+
"learning_rate": 1.6023408857646953e-05,
|
| 584 |
+
"loss": 4.5235,
|
| 585 |
+
"step": 3900
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 36.79,
|
| 589 |
+
"eval_loss": 3.035074472427368,
|
| 590 |
+
"eval_runtime": 2.4583,
|
| 591 |
+
"eval_samples_per_second": 291.661,
|
| 592 |
+
"eval_steps_per_second": 4.881,
|
| 593 |
+
"eval_wer": 0.10193526860193527,
|
| 594 |
+
"step": 3900
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 37.74,
|
| 598 |
+
"learning_rate": 1.575732466116854e-05,
|
| 599 |
+
"loss": 4.7715,
|
| 600 |
+
"step": 4000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 37.74,
|
| 604 |
+
"eval_loss": 3.023735761642456,
|
| 605 |
+
"eval_runtime": 2.3108,
|
| 606 |
+
"eval_samples_per_second": 310.281,
|
| 607 |
+
"eval_steps_per_second": 5.193,
|
| 608 |
+
"eval_wer": 0.09275942609275943,
|
| 609 |
+
"step": 4000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 38.68,
|
| 613 |
+
"learning_rate": 1.548499759213552e-05,
|
| 614 |
+
"loss": 4.7101,
|
| 615 |
+
"step": 4100
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 38.68,
|
| 619 |
+
"eval_loss": 3.024998426437378,
|
| 620 |
+
"eval_runtime": 2.2347,
|
| 621 |
+
"eval_samples_per_second": 320.853,
|
| 622 |
+
"eval_steps_per_second": 5.37,
|
| 623 |
+
"eval_wer": 0.09426092759426093,
|
| 624 |
+
"step": 4100
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 39.62,
|
| 628 |
+
"learning_rate": 1.5206722944505621e-05,
|
| 629 |
+
"loss": 4.243,
|
| 630 |
+
"step": 4200
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 39.62,
|
| 634 |
+
"eval_loss": 2.9704301357269287,
|
| 635 |
+
"eval_runtime": 2.278,
|
| 636 |
+
"eval_samples_per_second": 314.754,
|
| 637 |
+
"eval_steps_per_second": 5.268,
|
| 638 |
+
"eval_wer": 0.09801468134801468,
|
| 639 |
+
"step": 4200
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 40.57,
|
| 643 |
+
"learning_rate": 1.4922802461408265e-05,
|
| 644 |
+
"loss": 4.4015,
|
| 645 |
+
"step": 4300
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 40.57,
|
| 649 |
+
"eval_loss": 2.9600486755371094,
|
| 650 |
+
"eval_runtime": 2.138,
|
| 651 |
+
"eval_samples_per_second": 335.367,
|
| 652 |
+
"eval_steps_per_second": 5.613,
|
| 653 |
+
"eval_wer": 0.08708708708708708,
|
| 654 |
+
"step": 4300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 41.51,
|
| 658 |
+
"learning_rate": 1.4633544007953716e-05,
|
| 659 |
+
"loss": 4.4545,
|
| 660 |
+
"step": 4400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 41.51,
|
| 664 |
+
"eval_loss": 2.980609655380249,
|
| 665 |
+
"eval_runtime": 2.3548,
|
| 666 |
+
"eval_samples_per_second": 304.478,
|
| 667 |
+
"eval_steps_per_second": 5.096,
|
| 668 |
+
"eval_wer": 0.08583583583583583,
|
| 669 |
+
"step": 4400
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 42.45,
|
| 673 |
+
"learning_rate": 1.4339261237403948e-05,
|
| 674 |
+
"loss": 4.662,
|
| 675 |
+
"step": 4500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 42.45,
|
| 679 |
+
"eval_loss": 2.9668173789978027,
|
| 680 |
+
"eval_runtime": 2.2963,
|
| 681 |
+
"eval_samples_per_second": 312.242,
|
| 682 |
+
"eval_steps_per_second": 5.226,
|
| 683 |
+
"eval_wer": 0.09693026359693026,
|
| 684 |
+
"step": 4500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 43.4,
|
| 688 |
+
"learning_rate": 1.4043285361283684e-05,
|
| 689 |
+
"loss": 4.0696,
|
| 690 |
+
"step": 4600
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 43.4,
|
| 694 |
+
"eval_loss": 2.9349417686462402,
|
| 695 |
+
"eval_runtime": 2.305,
|
| 696 |
+
"eval_samples_per_second": 311.064,
|
| 697 |
+
"eval_steps_per_second": 5.206,
|
| 698 |
+
"eval_wer": 0.09351017684351018,
|
| 699 |
+
"step": 4600
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 44.34,
|
| 703 |
+
"learning_rate": 1.3739958551927287e-05,
|
| 704 |
+
"loss": 3.5668,
|
| 705 |
+
"step": 4700
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 44.34,
|
| 709 |
+
"eval_loss": 2.9190385341644287,
|
| 710 |
+
"eval_runtime": 2.3277,
|
| 711 |
+
"eval_samples_per_second": 308.023,
|
| 712 |
+
"eval_steps_per_second": 5.155,
|
| 713 |
+
"eval_wer": 0.09167500834167501,
|
| 714 |
+
"step": 4700
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 45.28,
|
| 718 |
+
"learning_rate": 1.3432576372069266e-05,
|
| 719 |
+
"loss": 3.8214,
|
| 720 |
+
"step": 4800
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 45.28,
|
| 724 |
+
"eval_loss": 2.948979139328003,
|
| 725 |
+
"eval_runtime": 2.1809,
|
| 726 |
+
"eval_samples_per_second": 328.766,
|
| 727 |
+
"eval_steps_per_second": 5.502,
|
| 728 |
+
"eval_wer": 0.09009009009009009,
|
| 729 |
+
"step": 4800
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 46.23,
|
| 733 |
+
"learning_rate": 1.3121472127174802e-05,
|
| 734 |
+
"loss": 3.8215,
|
| 735 |
+
"step": 4900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 46.23,
|
| 739 |
+
"eval_loss": 2.937129497528076,
|
| 740 |
+
"eval_runtime": 2.441,
|
| 741 |
+
"eval_samples_per_second": 293.729,
|
| 742 |
+
"eval_steps_per_second": 4.916,
|
| 743 |
+
"eval_wer": 0.0911745078411745,
|
| 744 |
+
"step": 4900
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 47.17,
|
| 748 |
+
"learning_rate": 1.2806983158677e-05,
|
| 749 |
+
"loss": 3.6593,
|
| 750 |
+
"step": 5000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 47.17,
|
| 754 |
+
"eval_loss": 2.9407706260681152,
|
| 755 |
+
"eval_runtime": 2.3032,
|
| 756 |
+
"eval_samples_per_second": 311.309,
|
| 757 |
+
"eval_steps_per_second": 5.21,
|
| 758 |
+
"eval_wer": 0.08750417083750417,
|
| 759 |
+
"step": 5000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 48.11,
|
| 763 |
+
"learning_rate": 1.248945047818555e-05,
|
| 764 |
+
"loss": 3.3709,
|
| 765 |
+
"step": 5100
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 48.11,
|
| 769 |
+
"eval_loss": 2.9576809406280518,
|
| 770 |
+
"eval_runtime": 2.3035,
|
| 771 |
+
"eval_samples_per_second": 311.269,
|
| 772 |
+
"eval_steps_per_second": 5.21,
|
| 773 |
+
"eval_wer": 0.09200867534200867,
|
| 774 |
+
"step": 5100
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 49.06,
|
| 778 |
+
"learning_rate": 1.2169218397715645e-05,
|
| 779 |
+
"loss": 3.5768,
|
| 780 |
+
"step": 5200
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 49.06,
|
| 784 |
+
"eval_loss": 2.986335039138794,
|
| 785 |
+
"eval_runtime": 2.3677,
|
| 786 |
+
"eval_samples_per_second": 302.826,
|
| 787 |
+
"eval_steps_per_second": 5.068,
|
| 788 |
+
"eval_wer": 0.09401067734401068,
|
| 789 |
+
"step": 5200
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 50.0,
|
| 793 |
+
"learning_rate": 1.1846634156338193e-05,
|
| 794 |
+
"loss": 3.3018,
|
| 795 |
+
"step": 5300
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 50.0,
|
| 799 |
+
"eval_loss": 2.943732500076294,
|
| 800 |
+
"eval_runtime": 2.3937,
|
| 801 |
+
"eval_samples_per_second": 299.537,
|
| 802 |
+
"eval_steps_per_second": 5.013,
|
| 803 |
+
"eval_wer": 0.10026693360026694,
|
| 804 |
+
"step": 5300
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 50.94,
|
| 808 |
+
"learning_rate": 1.1522047543656106e-05,
|
| 809 |
+
"loss": 3.2921,
|
| 810 |
+
"step": 5400
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 50.94,
|
| 814 |
+
"eval_loss": 2.9195058345794678,
|
| 815 |
+
"eval_runtime": 2.5086,
|
| 816 |
+
"eval_samples_per_second": 285.812,
|
| 817 |
+
"eval_steps_per_second": 4.783,
|
| 818 |
+
"eval_wer": 0.09225892559225893,
|
| 819 |
+
"step": 5400
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 51.89,
|
| 823 |
+
"learning_rate": 1.119581052051494e-05,
|
| 824 |
+
"loss": 3.4551,
|
| 825 |
+
"step": 5500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 51.89,
|
| 829 |
+
"eval_loss": 2.940958261489868,
|
| 830 |
+
"eval_runtime": 2.5071,
|
| 831 |
+
"eval_samples_per_second": 285.992,
|
| 832 |
+
"eval_steps_per_second": 4.786,
|
| 833 |
+
"eval_wer": 0.09501167834501167,
|
| 834 |
+
"step": 5500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 52.83,
|
| 838 |
+
"learning_rate": 1.0868276837359204e-05,
|
| 839 |
+
"loss": 3.6576,
|
| 840 |
+
"step": 5600
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 52.83,
|
| 844 |
+
"eval_loss": 2.9519896507263184,
|
| 845 |
+
"eval_runtime": 2.1953,
|
| 846 |
+
"eval_samples_per_second": 326.602,
|
| 847 |
+
"eval_steps_per_second": 5.466,
|
| 848 |
+
"eval_wer": 0.1011011011011011,
|
| 849 |
+
"step": 5600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 53.77,
|
| 853 |
+
"learning_rate": 1.0539801650648128e-05,
|
| 854 |
+
"loss": 3.5078,
|
| 855 |
+
"step": 5700
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 53.77,
|
| 859 |
+
"eval_loss": 2.8925814628601074,
|
| 860 |
+
"eval_runtime": 2.2343,
|
| 861 |
+
"eval_samples_per_second": 320.899,
|
| 862 |
+
"eval_steps_per_second": 5.371,
|
| 863 |
+
"eval_wer": 0.09367701034367701,
|
| 864 |
+
"step": 5700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 54.72,
|
| 868 |
+
"learning_rate": 1.0210741137746833e-05,
|
| 869 |
+
"loss": 3.0777,
|
| 870 |
+
"step": 5800
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 54.72,
|
| 874 |
+
"eval_loss": 2.8970932960510254,
|
| 875 |
+
"eval_runtime": 2.2598,
|
| 876 |
+
"eval_samples_per_second": 317.283,
|
| 877 |
+
"eval_steps_per_second": 5.31,
|
| 878 |
+
"eval_wer": 0.09134134134134134,
|
| 879 |
+
"step": 5800
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 55.66,
|
| 883 |
+
"learning_rate": 9.881452110710519e-06,
|
| 884 |
+
"loss": 3.0572,
|
| 885 |
+
"step": 5900
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 55.66,
|
| 889 |
+
"eval_loss": 2.8692612648010254,
|
| 890 |
+
"eval_runtime": 2.2528,
|
| 891 |
+
"eval_samples_per_second": 318.271,
|
| 892 |
+
"eval_steps_per_second": 5.327,
|
| 893 |
+
"eval_wer": 0.0890890890890891,
|
| 894 |
+
"step": 5900
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 56.6,
|
| 898 |
+
"learning_rate": 9.552291629380419e-06,
|
| 899 |
+
"loss": 3.0486,
|
| 900 |
+
"step": 6000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 56.6,
|
| 904 |
+
"eval_loss": 2.8876469135284424,
|
| 905 |
+
"eval_runtime": 2.2032,
|
| 906 |
+
"eval_samples_per_second": 325.435,
|
| 907 |
+
"eval_steps_per_second": 5.447,
|
| 908 |
+
"eval_wer": 0.0881715048381715,
|
| 909 |
+
"step": 6000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 57.55,
|
| 913 |
+
"learning_rate": 9.223616614211057e-06,
|
| 914 |
+
"loss": 3.1283,
|
| 915 |
+
"step": 6100
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 57.55,
|
| 919 |
+
"eval_loss": 2.8596761226654053,
|
| 920 |
+
"eval_runtime": 2.2244,
|
| 921 |
+
"eval_samples_per_second": 322.333,
|
| 922 |
+
"eval_steps_per_second": 5.395,
|
| 923 |
+
"eval_wer": 0.09134134134134134,
|
| 924 |
+
"step": 6100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 58.49,
|
| 928 |
+
"learning_rate": 8.895783459248695e-06,
|
| 929 |
+
"loss": 2.8705,
|
| 930 |
+
"step": 6200
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 58.49,
|
| 934 |
+
"eval_loss": 2.9079818725585938,
|
| 935 |
+
"eval_runtime": 2.3678,
|
| 936 |
+
"eval_samples_per_second": 302.819,
|
| 937 |
+
"eval_steps_per_second": 5.068,
|
| 938 |
+
"eval_wer": 0.09042375709042376,
|
| 939 |
+
"step": 6200
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 59.43,
|
| 943 |
+
"learning_rate": 8.569147645680523e-06,
|
| 944 |
+
"loss": 3.0644,
|
| 945 |
+
"step": 6300
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 59.43,
|
| 949 |
+
"eval_loss": 2.910602331161499,
|
| 950 |
+
"eval_runtime": 2.3829,
|
| 951 |
+
"eval_samples_per_second": 300.896,
|
| 952 |
+
"eval_steps_per_second": 5.036,
|
| 953 |
+
"eval_wer": 0.09167500834167501,
|
| 954 |
+
"step": 6300
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 60.38,
|
| 958 |
+
"learning_rate": 8.244063356373729e-06,
|
| 959 |
+
"loss": 2.8822,
|
| 960 |
+
"step": 6400
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 60.38,
|
| 964 |
+
"eval_loss": 2.9231059551239014,
|
| 965 |
+
"eval_runtime": 2.4556,
|
| 966 |
+
"eval_samples_per_second": 291.984,
|
| 967 |
+
"eval_steps_per_second": 4.887,
|
| 968 |
+
"eval_wer": 0.0890890890890891,
|
| 969 |
+
"step": 6400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 61.32,
|
| 973 |
+
"learning_rate": 7.92088309182241e-06,
|
| 974 |
+
"loss": 3.2338,
|
| 975 |
+
"step": 6500
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 61.32,
|
| 979 |
+
"eval_loss": 2.9510579109191895,
|
| 980 |
+
"eval_runtime": 2.1932,
|
| 981 |
+
"eval_samples_per_second": 326.917,
|
| 982 |
+
"eval_steps_per_second": 5.471,
|
| 983 |
+
"eval_wer": 0.09034034034034034,
|
| 984 |
+
"step": 6500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 62.26,
|
| 988 |
+
"learning_rate": 7.599957287918655e-06,
|
| 989 |
+
"loss": 3.048,
|
| 990 |
+
"step": 6600
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 62.26,
|
| 994 |
+
"eval_loss": 2.953885316848755,
|
| 995 |
+
"eval_runtime": 2.3592,
|
| 996 |
+
"eval_samples_per_second": 303.918,
|
| 997 |
+
"eval_steps_per_second": 5.086,
|
| 998 |
+
"eval_wer": 0.08975642308975643,
|
| 999 |
+
"step": 6600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 63.21,
|
| 1003 |
+
"learning_rate": 7.281633935962448e-06,
|
| 1004 |
+
"loss": 3.094,
|
| 1005 |
+
"step": 6700
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 63.21,
|
| 1009 |
+
"eval_loss": 2.949049949645996,
|
| 1010 |
+
"eval_runtime": 2.5261,
|
| 1011 |
+
"eval_samples_per_second": 283.836,
|
| 1012 |
+
"eval_steps_per_second": 4.75,
|
| 1013 |
+
"eval_wer": 0.09084084084084085,
|
| 1014 |
+
"step": 6700
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 64.15,
|
| 1018 |
+
"learning_rate": 6.966258205322274e-06,
|
| 1019 |
+
"loss": 3.0581,
|
| 1020 |
+
"step": 6800
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 64.15,
|
| 1024 |
+
"eval_loss": 2.895172119140625,
|
| 1025 |
+
"eval_runtime": 1.5993,
|
| 1026 |
+
"eval_samples_per_second": 448.308,
|
| 1027 |
+
"eval_steps_per_second": 7.503,
|
| 1028 |
+
"eval_wer": 0.08858858858858859,
|
| 1029 |
+
"step": 6800
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 65.09,
|
| 1033 |
+
"learning_rate": 6.654172069155635e-06,
|
| 1034 |
+
"loss": 2.9343,
|
| 1035 |
+
"step": 6900
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 65.09,
|
| 1039 |
+
"eval_loss": 2.8925557136535645,
|
| 1040 |
+
"eval_runtime": 2.4073,
|
| 1041 |
+
"eval_samples_per_second": 297.843,
|
| 1042 |
+
"eval_steps_per_second": 4.985,
|
| 1043 |
+
"eval_wer": 0.08825492158825492,
|
| 1044 |
+
"step": 6900
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 66.04,
|
| 1048 |
+
"learning_rate": 6.34571393359539e-06,
|
| 1049 |
+
"loss": 2.9497,
|
| 1050 |
+
"step": 7000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 66.04,
|
| 1054 |
+
"eval_loss": 2.8731818199157715,
|
| 1055 |
+
"eval_runtime": 2.4257,
|
| 1056 |
+
"eval_samples_per_second": 295.589,
|
| 1057 |
+
"eval_steps_per_second": 4.947,
|
| 1058 |
+
"eval_wer": 0.08875542208875542,
|
| 1059 |
+
"step": 7000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 66.98,
|
| 1063 |
+
"learning_rate": 6.0412182708038794e-06,
|
| 1064 |
+
"loss": 2.7788,
|
| 1065 |
+
"step": 7100
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 66.98,
|
| 1069 |
+
"eval_loss": 2.8836898803710938,
|
| 1070 |
+
"eval_runtime": 2.3112,
|
| 1071 |
+
"eval_samples_per_second": 310.231,
|
| 1072 |
+
"eval_steps_per_second": 5.192,
|
| 1073 |
+
"eval_wer": 0.09042375709042376,
|
| 1074 |
+
"step": 7100
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 67.92,
|
| 1078 |
+
"learning_rate": 5.741015256292837e-06,
|
| 1079 |
+
"loss": 2.7765,
|
| 1080 |
+
"step": 7200
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 67.92,
|
| 1084 |
+
"eval_loss": 2.9168927669525146,
|
| 1085 |
+
"eval_runtime": 2.2787,
|
| 1086 |
+
"eval_samples_per_second": 314.657,
|
| 1087 |
+
"eval_steps_per_second": 5.266,
|
| 1088 |
+
"eval_wer": 0.09509509509509509,
|
| 1089 |
+
"step": 7200
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 68.87,
|
| 1093 |
+
"learning_rate": 5.445430410902306e-06,
|
| 1094 |
+
"loss": 3.134,
|
| 1095 |
+
"step": 7300
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 68.87,
|
| 1099 |
+
"eval_loss": 2.903002977371216,
|
| 1100 |
+
"eval_runtime": 2.3536,
|
| 1101 |
+
"eval_samples_per_second": 304.633,
|
| 1102 |
+
"eval_steps_per_second": 5.098,
|
| 1103 |
+
"eval_wer": 0.09259259259259259,
|
| 1104 |
+
"step": 7300
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 69.81,
|
| 1108 |
+
"learning_rate": 5.15478424782678e-06,
|
| 1109 |
+
"loss": 2.8812,
|
| 1110 |
+
"step": 7400
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 69.81,
|
| 1114 |
+
"eval_loss": 2.9044971466064453,
|
| 1115 |
+
"eval_runtime": 2.2556,
|
| 1116 |
+
"eval_samples_per_second": 317.872,
|
| 1117 |
+
"eval_steps_per_second": 5.32,
|
| 1118 |
+
"eval_wer": 0.0920920920920921,
|
| 1119 |
+
"step": 7400
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 70.75,
|
| 1123 |
+
"learning_rate": 4.869391925071285e-06,
|
| 1124 |
+
"loss": 2.615,
|
| 1125 |
+
"step": 7500
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 70.75,
|
| 1129 |
+
"eval_loss": 2.9148054122924805,
|
| 1130 |
+
"eval_runtime": 2.2984,
|
| 1131 |
+
"eval_samples_per_second": 311.958,
|
| 1132 |
+
"eval_steps_per_second": 5.221,
|
| 1133 |
+
"eval_wer": 0.08708708708708708,
|
| 1134 |
+
"step": 7500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 71.7,
|
| 1138 |
+
"learning_rate": 4.589562903714348e-06,
|
| 1139 |
+
"loss": 2.5678,
|
| 1140 |
+
"step": 7600
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 71.7,
|
| 1144 |
+
"eval_loss": 2.943516969680786,
|
| 1145 |
+
"eval_runtime": 2.264,
|
| 1146 |
+
"eval_samples_per_second": 316.693,
|
| 1147 |
+
"eval_steps_per_second": 5.3,
|
| 1148 |
+
"eval_wer": 0.09217550884217551,
|
| 1149 |
+
"step": 7600
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 72.64,
|
| 1153 |
+
"learning_rate": 4.318310214093595e-06,
|
| 1154 |
+
"loss": 2.4858,
|
| 1155 |
+
"step": 7700
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 72.64,
|
| 1159 |
+
"eval_loss": 2.9050374031066895,
|
| 1160 |
+
"eval_runtime": 2.5001,
|
| 1161 |
+
"eval_samples_per_second": 286.789,
|
| 1162 |
+
"eval_steps_per_second": 4.8,
|
| 1163 |
+
"eval_wer": 0.09284284284284285,
|
| 1164 |
+
"step": 7700
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 73.58,
|
| 1168 |
+
"learning_rate": 4.050448633035326e-06,
|
| 1169 |
+
"loss": 2.5367,
|
| 1170 |
+
"step": 7800
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 73.58,
|
| 1174 |
+
"eval_loss": 2.8948252201080322,
|
| 1175 |
+
"eval_runtime": 2.3819,
|
| 1176 |
+
"eval_samples_per_second": 301.016,
|
| 1177 |
+
"eval_steps_per_second": 5.038,
|
| 1178 |
+
"eval_wer": 0.08775442108775443,
|
| 1179 |
+
"step": 7800
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 74.53,
|
| 1183 |
+
"learning_rate": 3.7890383628075156e-06,
|
| 1184 |
+
"loss": 2.3228,
|
| 1185 |
+
"step": 7900
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 74.53,
|
| 1189 |
+
"eval_loss": 2.8994953632354736,
|
| 1190 |
+
"eval_runtime": 2.4234,
|
| 1191 |
+
"eval_samples_per_second": 295.866,
|
| 1192 |
+
"eval_steps_per_second": 4.952,
|
| 1193 |
+
"eval_wer": 0.0890890890890891,
|
| 1194 |
+
"step": 7900
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 75.47,
|
| 1198 |
+
"learning_rate": 3.5343628598934275e-06,
|
| 1199 |
+
"loss": 2.5849,
|
| 1200 |
+
"step": 8000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 75.47,
|
| 1204 |
+
"eval_loss": 2.9289326667785645,
|
| 1205 |
+
"eval_runtime": 2.4566,
|
| 1206 |
+
"eval_samples_per_second": 291.864,
|
| 1207 |
+
"eval_steps_per_second": 4.885,
|
| 1208 |
+
"eval_wer": 0.09275942609275943,
|
| 1209 |
+
"step": 8000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 76.42,
|
| 1213 |
+
"learning_rate": 3.2866982780278357e-06,
|
| 1214 |
+
"loss": 2.6645,
|
| 1215 |
+
"step": 8100
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 76.42,
|
| 1219 |
+
"eval_loss": 2.8950443267822266,
|
| 1220 |
+
"eval_runtime": 2.3369,
|
| 1221 |
+
"eval_samples_per_second": 306.813,
|
| 1222 |
+
"eval_steps_per_second": 5.135,
|
| 1223 |
+
"eval_wer": 0.08842175508842176,
|
| 1224 |
+
"step": 8100
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 77.36,
|
| 1228 |
+
"learning_rate": 3.0463131687536695e-06,
|
| 1229 |
+
"loss": 2.6634,
|
| 1230 |
+
"step": 8200
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 77.36,
|
| 1234 |
+
"eval_loss": 2.919368267059326,
|
| 1235 |
+
"eval_runtime": 2.2121,
|
| 1236 |
+
"eval_samples_per_second": 324.122,
|
| 1237 |
+
"eval_steps_per_second": 5.425,
|
| 1238 |
+
"eval_wer": 0.09217550884217551,
|
| 1239 |
+
"step": 8200
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 78.3,
|
| 1243 |
+
"learning_rate": 2.8134681902219797e-06,
|
| 1244 |
+
"loss": 2.393,
|
| 1245 |
+
"step": 8300
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 78.3,
|
| 1249 |
+
"eval_loss": 2.9073572158813477,
|
| 1250 |
+
"eval_runtime": 2.3983,
|
| 1251 |
+
"eval_samples_per_second": 298.96,
|
| 1252 |
+
"eval_steps_per_second": 5.004,
|
| 1253 |
+
"eval_wer": 0.09192525859192525,
|
| 1254 |
+
"step": 8300
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 79.25,
|
| 1258 |
+
"learning_rate": 2.5884158245510325e-06,
|
| 1259 |
+
"loss": 3.0675,
|
| 1260 |
+
"step": 8400
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 79.25,
|
| 1264 |
+
"eval_loss": 2.892658233642578,
|
| 1265 |
+
"eval_runtime": 1.8302,
|
| 1266 |
+
"eval_samples_per_second": 391.764,
|
| 1267 |
+
"eval_steps_per_second": 6.557,
|
| 1268 |
+
"eval_wer": 0.09075742409075742,
|
| 1269 |
+
"step": 8400
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 80.19,
|
| 1273 |
+
"learning_rate": 2.3714001040509095e-06,
|
| 1274 |
+
"loss": 2.6344,
|
| 1275 |
+
"step": 8500
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 80.19,
|
| 1279 |
+
"eval_loss": 2.876831293106079,
|
| 1280 |
+
"eval_runtime": 1.6456,
|
| 1281 |
+
"eval_samples_per_second": 435.704,
|
| 1282 |
+
"eval_steps_per_second": 7.292,
|
| 1283 |
+
"eval_wer": 0.0890890890890891,
|
| 1284 |
+
"step": 8500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 81.13,
|
| 1288 |
+
"learning_rate": 2.162656346610592e-06,
|
| 1289 |
+
"loss": 2.5742,
|
| 1290 |
+
"step": 8600
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 81.13,
|
| 1294 |
+
"eval_loss": 2.8808791637420654,
|
| 1295 |
+
"eval_runtime": 2.9484,
|
| 1296 |
+
"eval_samples_per_second": 243.18,
|
| 1297 |
+
"eval_steps_per_second": 4.07,
|
| 1298 |
+
"eval_wer": 0.09109109109109109,
|
| 1299 |
+
"step": 8600
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 82.08,
|
| 1303 |
+
"learning_rate": 1.962410900534367e-06,
|
| 1304 |
+
"loss": 2.6523,
|
| 1305 |
+
"step": 8700
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 82.08,
|
| 1309 |
+
"eval_loss": 2.863851547241211,
|
| 1310 |
+
"eval_runtime": 2.6818,
|
| 1311 |
+
"eval_samples_per_second": 267.357,
|
| 1312 |
+
"eval_steps_per_second": 4.475,
|
| 1313 |
+
"eval_wer": 0.08625291958625292,
|
| 1314 |
+
"step": 8700
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 83.02,
|
| 1318 |
+
"learning_rate": 1.7746247286710604e-06,
|
| 1319 |
+
"loss": 2.2657,
|
| 1320 |
+
"step": 8800
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 83.02,
|
| 1324 |
+
"eval_loss": 2.8809173107147217,
|
| 1325 |
+
"eval_runtime": 2.6706,
|
| 1326 |
+
"eval_samples_per_second": 268.481,
|
| 1327 |
+
"eval_steps_per_second": 4.493,
|
| 1328 |
+
"eval_wer": 0.0911745078411745,
|
| 1329 |
+
"step": 8800
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 83.96,
|
| 1333 |
+
"learning_rate": 1.591837419081773e-06,
|
| 1334 |
+
"loss": 2.3238,
|
| 1335 |
+
"step": 8900
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 83.96,
|
| 1339 |
+
"eval_loss": 2.876391887664795,
|
| 1340 |
+
"eval_runtime": 2.5394,
|
| 1341 |
+
"eval_samples_per_second": 282.355,
|
| 1342 |
+
"eval_steps_per_second": 4.726,
|
| 1343 |
+
"eval_wer": 0.08933933933933934,
|
| 1344 |
+
"step": 8900
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 84.91,
|
| 1348 |
+
"learning_rate": 1.4181673801868301e-06,
|
| 1349 |
+
"loss": 2.3664,
|
| 1350 |
+
"step": 9000
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 84.91,
|
| 1354 |
+
"eval_loss": 2.873842239379883,
|
| 1355 |
+
"eval_runtime": 2.4749,
|
| 1356 |
+
"eval_samples_per_second": 289.712,
|
| 1357 |
+
"eval_steps_per_second": 4.849,
|
| 1358 |
+
"eval_wer": 0.09125792459125792,
|
| 1359 |
+
"step": 9000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 85.85,
|
| 1363 |
+
"learning_rate": 1.2538029286060428e-06,
|
| 1364 |
+
"loss": 2.5655,
|
| 1365 |
+
"step": 9100
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 85.85,
|
| 1369 |
+
"eval_loss": 2.887613296508789,
|
| 1370 |
+
"eval_runtime": 2.3857,
|
| 1371 |
+
"eval_samples_per_second": 300.547,
|
| 1372 |
+
"eval_steps_per_second": 5.03,
|
| 1373 |
+
"eval_wer": 0.09042375709042376,
|
| 1374 |
+
"step": 9100
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 86.79,
|
| 1378 |
+
"learning_rate": 1.0989222905788767e-06,
|
| 1379 |
+
"loss": 2.4372,
|
| 1380 |
+
"step": 9200
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 86.79,
|
| 1384 |
+
"eval_loss": 2.902373790740967,
|
| 1385 |
+
"eval_runtime": 2.3648,
|
| 1386 |
+
"eval_samples_per_second": 303.2,
|
| 1387 |
+
"eval_steps_per_second": 5.074,
|
| 1388 |
+
"eval_wer": 0.09100767434100768,
|
| 1389 |
+
"step": 9200
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 87.74,
|
| 1393 |
+
"learning_rate": 9.536934087073702e-07,
|
| 1394 |
+
"loss": 2.5267,
|
| 1395 |
+
"step": 9300
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 87.74,
|
| 1399 |
+
"eval_loss": 2.892174243927002,
|
| 1400 |
+
"eval_runtime": 2.5482,
|
| 1401 |
+
"eval_samples_per_second": 281.372,
|
| 1402 |
+
"eval_steps_per_second": 4.709,
|
| 1403 |
+
"eval_wer": 0.08975642308975643,
|
| 1404 |
+
"step": 9300
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 88.68,
|
| 1408 |
+
"learning_rate": 8.182737598499846e-07,
|
| 1409 |
+
"loss": 2.471,
|
| 1410 |
+
"step": 9400
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 88.68,
|
| 1414 |
+
"eval_loss": 2.889312744140625,
|
| 1415 |
+
"eval_runtime": 2.3375,
|
| 1416 |
+
"eval_samples_per_second": 306.735,
|
| 1417 |
+
"eval_steps_per_second": 5.134,
|
| 1418 |
+
"eval_wer": 0.08842175508842176,
|
| 1419 |
+
"step": 9400
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 89.62,
|
| 1423 |
+
"learning_rate": 6.928101843638202e-07,
|
| 1424 |
+
"loss": 2.5225,
|
| 1425 |
+
"step": 9500
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 89.62,
|
| 1429 |
+
"eval_loss": 2.885225296020508,
|
| 1430 |
+
"eval_runtime": 2.3766,
|
| 1431 |
+
"eval_samples_per_second": 301.687,
|
| 1432 |
+
"eval_steps_per_second": 5.049,
|
| 1433 |
+
"eval_wer": 0.08875542208875542,
|
| 1434 |
+
"step": 9500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 90.57,
|
| 1438 |
+
"learning_rate": 5.774387268803871e-07,
|
| 1439 |
+
"loss": 2.4752,
|
| 1440 |
+
"step": 9600
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 90.57,
|
| 1444 |
+
"eval_loss": 2.8876028060913086,
|
| 1445 |
+
"eval_runtime": 2.4035,
|
| 1446 |
+
"eval_samples_per_second": 298.319,
|
| 1447 |
+
"eval_steps_per_second": 4.993,
|
| 1448 |
+
"eval_wer": 0.08917250583917251,
|
| 1449 |
+
"step": 9600
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 91.51,
|
| 1453 |
+
"learning_rate": 4.722844887875522e-07,
|
| 1454 |
+
"loss": 2.5029,
|
| 1455 |
+
"step": 9700
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 91.51,
|
| 1459 |
+
"eval_loss": 2.8883309364318848,
|
| 1460 |
+
"eval_runtime": 2.5391,
|
| 1461 |
+
"eval_samples_per_second": 282.383,
|
| 1462 |
+
"eval_steps_per_second": 4.726,
|
| 1463 |
+
"eval_wer": 0.08850517183850518,
|
| 1464 |
+
"step": 9700
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 92.45,
|
| 1468 |
+
"learning_rate": 3.7746149257763984e-07,
|
| 1469 |
+
"loss": 2.7052,
|
| 1470 |
+
"step": 9800
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 92.45,
|
| 1474 |
+
"eval_loss": 2.882530450820923,
|
| 1475 |
+
"eval_runtime": 2.5114,
|
| 1476 |
+
"eval_samples_per_second": 285.5,
|
| 1477 |
+
"eval_steps_per_second": 4.778,
|
| 1478 |
+
"eval_wer": 0.08708708708708708,
|
| 1479 |
+
"step": 9800
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 93.4,
|
| 1483 |
+
"learning_rate": 2.9307255820877676e-07,
|
| 1484 |
+
"loss": 2.4682,
|
| 1485 |
+
"step": 9900
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 93.4,
|
| 1489 |
+
"eval_loss": 2.8780291080474854,
|
| 1490 |
+
"eval_runtime": 2.3878,
|
| 1491 |
+
"eval_samples_per_second": 300.275,
|
| 1492 |
+
"eval_steps_per_second": 5.026,
|
| 1493 |
+
"eval_wer": 0.08700367033700367,
|
| 1494 |
+
"step": 9900
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 94.34,
|
| 1498 |
+
"learning_rate": 2.1920919161354304e-07,
|
| 1499 |
+
"loss": 2.3672,
|
| 1500 |
+
"step": 10000
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 94.34,
|
| 1504 |
+
"eval_loss": 2.8809967041015625,
|
| 1505 |
+
"eval_runtime": 2.5256,
|
| 1506 |
+
"eval_samples_per_second": 283.891,
|
| 1507 |
+
"eval_steps_per_second": 4.751,
|
| 1508 |
+
"eval_wer": 0.08717050383717051,
|
| 1509 |
+
"step": 10000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 95.28,
|
| 1513 |
+
"learning_rate": 1.5595148547582373e-07,
|
| 1514 |
+
"loss": 2.5325,
|
| 1515 |
+
"step": 10100
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 95.28,
|
| 1519 |
+
"eval_loss": 2.88421630859375,
|
| 1520 |
+
"eval_runtime": 2.5311,
|
| 1521 |
+
"eval_samples_per_second": 283.279,
|
| 1522 |
+
"eval_steps_per_second": 4.741,
|
| 1523 |
+
"eval_wer": 0.08842175508842176,
|
| 1524 |
+
"step": 10100
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 96.23,
|
| 1528 |
+
"learning_rate": 1.0336803238345072e-07,
|
| 1529 |
+
"loss": 2.4877,
|
| 1530 |
+
"step": 10200
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 96.23,
|
| 1534 |
+
"eval_loss": 2.88330340385437,
|
| 1535 |
+
"eval_runtime": 2.3205,
|
| 1536 |
+
"eval_samples_per_second": 308.982,
|
| 1537 |
+
"eval_steps_per_second": 5.171,
|
| 1538 |
+
"eval_wer": 0.08842175508842176,
|
| 1539 |
+
"step": 10200
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 97.17,
|
| 1543 |
+
"learning_rate": 6.151585045082286e-08,
|
| 1544 |
+
"loss": 2.7373,
|
| 1545 |
+
"step": 10300
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 97.17,
|
| 1549 |
+
"eval_loss": 2.8824727535247803,
|
| 1550 |
+
"eval_runtime": 2.4169,
|
| 1551 |
+
"eval_samples_per_second": 296.665,
|
| 1552 |
+
"eval_steps_per_second": 4.965,
|
| 1553 |
+
"eval_wer": 0.0881715048381715,
|
| 1554 |
+
"step": 10300
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 98.11,
|
| 1558 |
+
"learning_rate": 3.044032149211562e-08,
|
| 1559 |
+
"loss": 2.5574,
|
| 1560 |
+
"step": 10400
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 98.11,
|
| 1564 |
+
"eval_loss": 2.8832902908325195,
|
| 1565 |
+
"eval_runtime": 2.3516,
|
| 1566 |
+
"eval_samples_per_second": 304.903,
|
| 1567 |
+
"eval_steps_per_second": 5.103,
|
| 1568 |
+
"eval_wer": 0.08808808808808809,
|
| 1569 |
+
"step": 10400
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 99.06,
|
| 1573 |
+
"learning_rate": 1.0175141812170941e-08,
|
| 1574 |
+
"loss": 2.2097,
|
| 1575 |
+
"step": 10500
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 99.06,
|
| 1579 |
+
"eval_loss": 2.8823037147521973,
|
| 1580 |
+
"eval_runtime": 2.2498,
|
| 1581 |
+
"eval_samples_per_second": 318.7,
|
| 1582 |
+
"eval_steps_per_second": 5.334,
|
| 1583 |
+
"eval_wer": 0.08825492158825492,
|
| 1584 |
+
"step": 10500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 100.0,
|
| 1588 |
+
"learning_rate": 7.422856683902968e-10,
|
| 1589 |
+
"loss": 2.5919,
|
| 1590 |
+
"step": 10600
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 100.0,
|
| 1594 |
+
"eval_loss": 2.8828866481781006,
|
| 1595 |
+
"eval_runtime": 2.2981,
|
| 1596 |
+
"eval_samples_per_second": 312.003,
|
| 1597 |
+
"eval_steps_per_second": 5.222,
|
| 1598 |
+
"eval_wer": 0.08792125458792126,
|
| 1599 |
+
"step": 10600
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 100.0,
|
| 1603 |
+
"step": 10600,
|
| 1604 |
+
"total_flos": 2.842046364754798e+19,
|
| 1605 |
+
"train_loss": 32.606080211423475,
|
| 1606 |
+
"train_runtime": 7534.6733,
|
| 1607 |
+
"train_samples_per_second": 44.912,
|
| 1608 |
+
"train_steps_per_second": 1.407
|
| 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 |
+
}
|