End of training
Browse files- trainer_state.json +1618 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1618 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 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": 119.415,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.94,
|
| 19 |
+
"eval_loss": 91.51124572753906,
|
| 20 |
+
"eval_runtime": 2.4593,
|
| 21 |
+
"eval_samples_per_second": 291.548,
|
| 22 |
+
"eval_steps_per_second": 4.879,
|
| 23 |
+
"eval_wer": 18.636386386386388,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.89,
|
| 28 |
+
"learning_rate": 3.2264150943396226e-06,
|
| 29 |
+
"loss": 74.7916,
|
| 30 |
+
"step": 200
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.89,
|
| 34 |
+
"eval_loss": 12.292832374572754,
|
| 35 |
+
"eval_runtime": 1.6185,
|
| 36 |
+
"eval_samples_per_second": 443.014,
|
| 37 |
+
"eval_steps_per_second": 7.414,
|
| 38 |
+
"eval_wer": 0.9950784117450784,
|
| 39 |
+
"step": 200
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.83,
|
| 43 |
+
"learning_rate": 5.1132075471698114e-06,
|
| 44 |
+
"loss": 6.9068,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.83,
|
| 49 |
+
"eval_loss": 5.234492301940918,
|
| 50 |
+
"eval_runtime": 1.5874,
|
| 51 |
+
"eval_samples_per_second": 451.676,
|
| 52 |
+
"eval_steps_per_second": 7.559,
|
| 53 |
+
"eval_wer": 1.0,
|
| 54 |
+
"step": 300
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.77,
|
| 58 |
+
"learning_rate": 7e-06,
|
| 59 |
+
"loss": 5.1207,
|
| 60 |
+
"step": 400
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 3.77,
|
| 64 |
+
"eval_loss": 5.036521911621094,
|
| 65 |
+
"eval_runtime": 1.6963,
|
| 66 |
+
"eval_samples_per_second": 422.679,
|
| 67 |
+
"eval_steps_per_second": 7.074,
|
| 68 |
+
"eval_wer": 1.0,
|
| 69 |
+
"step": 400
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 4.72,
|
| 73 |
+
"learning_rate": 8.88679245283019e-06,
|
| 74 |
+
"loss": 4.7306,
|
| 75 |
+
"step": 500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 4.72,
|
| 79 |
+
"eval_loss": 4.915191650390625,
|
| 80 |
+
"eval_runtime": 1.5665,
|
| 81 |
+
"eval_samples_per_second": 457.698,
|
| 82 |
+
"eval_steps_per_second": 7.66,
|
| 83 |
+
"eval_wer": 1.0,
|
| 84 |
+
"step": 500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.66,
|
| 88 |
+
"learning_rate": 1.077358490566038e-05,
|
| 89 |
+
"loss": 4.4974,
|
| 90 |
+
"step": 600
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 5.66,
|
| 94 |
+
"eval_loss": 4.9314799308776855,
|
| 95 |
+
"eval_runtime": 1.6678,
|
| 96 |
+
"eval_samples_per_second": 429.898,
|
| 97 |
+
"eval_steps_per_second": 7.195,
|
| 98 |
+
"eval_wer": 1.0,
|
| 99 |
+
"step": 600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.6,
|
| 103 |
+
"learning_rate": 1.2660377358490568e-05,
|
| 104 |
+
"loss": 4.3923,
|
| 105 |
+
"step": 700
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.6,
|
| 109 |
+
"eval_loss": 4.791795253753662,
|
| 110 |
+
"eval_runtime": 1.6083,
|
| 111 |
+
"eval_samples_per_second": 445.801,
|
| 112 |
+
"eval_steps_per_second": 7.461,
|
| 113 |
+
"eval_wer": 1.0,
|
| 114 |
+
"step": 700
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.55,
|
| 118 |
+
"learning_rate": 1.4547169811320756e-05,
|
| 119 |
+
"loss": 4.3447,
|
| 120 |
+
"step": 800
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.55,
|
| 124 |
+
"eval_loss": 4.644739627838135,
|
| 125 |
+
"eval_runtime": 1.551,
|
| 126 |
+
"eval_samples_per_second": 462.283,
|
| 127 |
+
"eval_steps_per_second": 7.737,
|
| 128 |
+
"eval_wer": 1.0,
|
| 129 |
+
"step": 800
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 8.49,
|
| 133 |
+
"learning_rate": 1.6433962264150943e-05,
|
| 134 |
+
"loss": 4.225,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.49,
|
| 139 |
+
"eval_loss": 4.6060686111450195,
|
| 140 |
+
"eval_runtime": 1.5063,
|
| 141 |
+
"eval_samples_per_second": 476.004,
|
| 142 |
+
"eval_steps_per_second": 7.967,
|
| 143 |
+
"eval_wer": 1.0,
|
| 144 |
+
"step": 900
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.43,
|
| 148 |
+
"learning_rate": 1.832075471698113e-05,
|
| 149 |
+
"loss": 3.9805,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.43,
|
| 154 |
+
"eval_loss": 3.642165422439575,
|
| 155 |
+
"eval_runtime": 1.5007,
|
| 156 |
+
"eval_samples_per_second": 477.771,
|
| 157 |
+
"eval_steps_per_second": 7.996,
|
| 158 |
+
"eval_wer": 0.87328995662329,
|
| 159 |
+
"step": 1000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 10.38,
|
| 163 |
+
"learning_rate": 1.99999343918343e-05,
|
| 164 |
+
"loss": 2.8303,
|
| 165 |
+
"step": 1100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 10.38,
|
| 169 |
+
"eval_loss": 1.782403588294983,
|
| 170 |
+
"eval_runtime": 1.6315,
|
| 171 |
+
"eval_samples_per_second": 439.485,
|
| 172 |
+
"eval_steps_per_second": 7.355,
|
| 173 |
+
"eval_wer": 0.34893226559893226,
|
| 174 |
+
"step": 1100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 11.32,
|
| 178 |
+
"learning_rate": 1.999332009015375e-05,
|
| 179 |
+
"loss": 1.5807,
|
| 180 |
+
"step": 1200
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 11.32,
|
| 184 |
+
"eval_loss": 1.090773105621338,
|
| 185 |
+
"eval_runtime": 1.5184,
|
| 186 |
+
"eval_samples_per_second": 472.209,
|
| 187 |
+
"eval_steps_per_second": 7.903,
|
| 188 |
+
"eval_wer": 0.21621621621621623,
|
| 189 |
+
"step": 1200
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 12.26,
|
| 193 |
+
"learning_rate": 1.9975869674932303e-05,
|
| 194 |
+
"loss": 1.1284,
|
| 195 |
+
"step": 1300
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.26,
|
| 199 |
+
"eval_loss": 0.8473044037818909,
|
| 200 |
+
"eval_runtime": 1.6911,
|
| 201 |
+
"eval_samples_per_second": 423.989,
|
| 202 |
+
"eval_steps_per_second": 7.096,
|
| 203 |
+
"eval_wer": 0.16399733066399733,
|
| 204 |
+
"step": 1300
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 13.21,
|
| 208 |
+
"learning_rate": 1.9947602068277822e-05,
|
| 209 |
+
"loss": 0.8703,
|
| 210 |
+
"step": 1400
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.21,
|
| 214 |
+
"eval_loss": 0.7322315573692322,
|
| 215 |
+
"eval_runtime": 1.5817,
|
| 216 |
+
"eval_samples_per_second": 453.303,
|
| 217 |
+
"eval_steps_per_second": 7.587,
|
| 218 |
+
"eval_wer": 0.142308975642309,
|
| 219 |
+
"step": 1400
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 14.15,
|
| 223 |
+
"learning_rate": 1.9908547921764814e-05,
|
| 224 |
+
"loss": 0.7576,
|
| 225 |
+
"step": 1500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 14.15,
|
| 229 |
+
"eval_loss": 0.6551467180252075,
|
| 230 |
+
"eval_runtime": 1.5088,
|
| 231 |
+
"eval_samples_per_second": 475.207,
|
| 232 |
+
"eval_steps_per_second": 7.953,
|
| 233 |
+
"eval_wer": 0.13254921588254923,
|
| 234 |
+
"step": 1500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 15.09,
|
| 238 |
+
"learning_rate": 1.985874958319781e-05,
|
| 239 |
+
"loss": 0.6256,
|
| 240 |
+
"step": 1600
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.09,
|
| 244 |
+
"eval_loss": 0.6027040481567383,
|
| 245 |
+
"eval_runtime": 1.5077,
|
| 246 |
+
"eval_samples_per_second": 475.558,
|
| 247 |
+
"eval_steps_per_second": 7.959,
|
| 248 |
+
"eval_wer": 0.13872205538872207,
|
| 249 |
+
"step": 1600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.04,
|
| 253 |
+
"learning_rate": 1.9798261050692154e-05,
|
| 254 |
+
"loss": 0.594,
|
| 255 |
+
"step": 1700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 16.04,
|
| 259 |
+
"eval_loss": 0.5549562573432922,
|
| 260 |
+
"eval_runtime": 1.5493,
|
| 261 |
+
"eval_samples_per_second": 462.792,
|
| 262 |
+
"eval_steps_per_second": 7.745,
|
| 263 |
+
"eval_wer": 0.13004671338004672,
|
| 264 |
+
"step": 1700
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.98,
|
| 268 |
+
"learning_rate": 1.9727147914121905e-05,
|
| 269 |
+
"loss": 0.5492,
|
| 270 |
+
"step": 1800
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.98,
|
| 274 |
+
"eval_loss": 0.5200316309928894,
|
| 275 |
+
"eval_runtime": 1.5968,
|
| 276 |
+
"eval_samples_per_second": 449.021,
|
| 277 |
+
"eval_steps_per_second": 7.515,
|
| 278 |
+
"eval_wer": 0.11594928261594928,
|
| 279 |
+
"step": 1800
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 17.92,
|
| 283 |
+
"learning_rate": 1.96454872839984e-05,
|
| 284 |
+
"loss": 0.476,
|
| 285 |
+
"step": 1900
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 17.92,
|
| 289 |
+
"eval_loss": 0.5012216567993164,
|
| 290 |
+
"eval_runtime": 1.6883,
|
| 291 |
+
"eval_samples_per_second": 424.678,
|
| 292 |
+
"eval_steps_per_second": 7.108,
|
| 293 |
+
"eval_wer": 0.1091091091091091,
|
| 294 |
+
"step": 1900
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 18.87,
|
| 298 |
+
"learning_rate": 1.9553367707856575e-05,
|
| 299 |
+
"loss": 0.4822,
|
| 300 |
+
"step": 2000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 18.87,
|
| 304 |
+
"eval_loss": 0.5111868977546692,
|
| 305 |
+
"eval_runtime": 1.579,
|
| 306 |
+
"eval_samples_per_second": 454.084,
|
| 307 |
+
"eval_steps_per_second": 7.6,
|
| 308 |
+
"eval_wer": 0.10735735735735735,
|
| 309 |
+
"step": 2000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 19.81,
|
| 313 |
+
"learning_rate": 1.9450889074239728e-05,
|
| 314 |
+
"loss": 0.4351,
|
| 315 |
+
"step": 2100
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 19.81,
|
| 319 |
+
"eval_loss": 0.4985233247280121,
|
| 320 |
+
"eval_runtime": 1.6359,
|
| 321 |
+
"eval_samples_per_second": 438.301,
|
| 322 |
+
"eval_steps_per_second": 7.336,
|
| 323 |
+
"eval_wer": 0.11786786786786786,
|
| 324 |
+
"step": 2100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 20.75,
|
| 328 |
+
"learning_rate": 1.9338162504386778e-05,
|
| 329 |
+
"loss": 0.4169,
|
| 330 |
+
"step": 2200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 20.75,
|
| 334 |
+
"eval_loss": 0.47119975090026855,
|
| 335 |
+
"eval_runtime": 1.6934,
|
| 336 |
+
"eval_samples_per_second": 423.414,
|
| 337 |
+
"eval_steps_per_second": 7.086,
|
| 338 |
+
"eval_wer": 0.1061061061061061,
|
| 339 |
+
"step": 2200
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 21.7,
|
| 343 |
+
"learning_rate": 1.9216588448323045e-05,
|
| 344 |
+
"loss": 0.3957,
|
| 345 |
+
"step": 2300
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 21.7,
|
| 349 |
+
"eval_loss": 0.461343377828598,
|
| 350 |
+
"eval_runtime": 1.5127,
|
| 351 |
+
"eval_samples_per_second": 473.993,
|
| 352 |
+
"eval_steps_per_second": 7.933,
|
| 353 |
+
"eval_wer": 0.09876543209876543,
|
| 354 |
+
"step": 2300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 22.64,
|
| 358 |
+
"learning_rate": 1.908384291577325e-05,
|
| 359 |
+
"loss": 0.3885,
|
| 360 |
+
"step": 2400
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 22.64,
|
| 364 |
+
"eval_loss": 0.4609617590904236,
|
| 365 |
+
"eval_runtime": 1.6911,
|
| 366 |
+
"eval_samples_per_second": 423.997,
|
| 367 |
+
"eval_steps_per_second": 7.096,
|
| 368 |
+
"eval_wer": 0.10251918585251918,
|
| 369 |
+
"step": 2400
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 23.58,
|
| 373 |
+
"learning_rate": 1.8941247448233386e-05,
|
| 374 |
+
"loss": 0.3827,
|
| 375 |
+
"step": 2500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 23.58,
|
| 379 |
+
"eval_loss": 0.4508582353591919,
|
| 380 |
+
"eval_runtime": 1.6027,
|
| 381 |
+
"eval_samples_per_second": 447.383,
|
| 382 |
+
"eval_steps_per_second": 7.488,
|
| 383 |
+
"eval_wer": 0.09784784784784785,
|
| 384 |
+
"step": 2500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.53,
|
| 388 |
+
"learning_rate": 1.878895666705679e-05,
|
| 389 |
+
"loss": 0.3468,
|
| 390 |
+
"step": 2600
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 24.53,
|
| 394 |
+
"eval_loss": 0.45491451025009155,
|
| 395 |
+
"eval_runtime": 1.5761,
|
| 396 |
+
"eval_samples_per_second": 454.906,
|
| 397 |
+
"eval_steps_per_second": 7.613,
|
| 398 |
+
"eval_wer": 0.09509509509509509,
|
| 399 |
+
"step": 2600
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 25.47,
|
| 403 |
+
"learning_rate": 1.8627135706571312e-05,
|
| 404 |
+
"loss": 0.3451,
|
| 405 |
+
"step": 2700
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 25.47,
|
| 409 |
+
"eval_loss": 0.4556306004524231,
|
| 410 |
+
"eval_runtime": 1.5067,
|
| 411 |
+
"eval_samples_per_second": 475.867,
|
| 412 |
+
"eval_steps_per_second": 7.964,
|
| 413 |
+
"eval_wer": 0.10185185185185185,
|
| 414 |
+
"step": 2700
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 26.42,
|
| 418 |
+
"learning_rate": 1.845596003501826e-05,
|
| 419 |
+
"loss": 0.3234,
|
| 420 |
+
"step": 2800
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 26.42,
|
| 424 |
+
"eval_loss": 0.4554463028907776,
|
| 425 |
+
"eval_runtime": 1.5115,
|
| 426 |
+
"eval_samples_per_second": 474.37,
|
| 427 |
+
"eval_steps_per_second": 7.939,
|
| 428 |
+
"eval_wer": 0.11036036036036036,
|
| 429 |
+
"step": 2800
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 27.36,
|
| 433 |
+
"learning_rate": 1.827746346507147e-05,
|
| 434 |
+
"loss": 0.31,
|
| 435 |
+
"step": 2900
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 27.36,
|
| 439 |
+
"eval_loss": 0.45678067207336426,
|
| 440 |
+
"eval_runtime": 1.6353,
|
| 441 |
+
"eval_samples_per_second": 438.463,
|
| 442 |
+
"eval_steps_per_second": 7.338,
|
| 443 |
+
"eval_wer": 0.09876543209876543,
|
| 444 |
+
"step": 2900
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 28.3,
|
| 448 |
+
"learning_rate": 1.808823388475444e-05,
|
| 449 |
+
"loss": 0.3026,
|
| 450 |
+
"step": 3000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 28.3,
|
| 454 |
+
"eval_loss": 0.42110905051231384,
|
| 455 |
+
"eval_runtime": 1.5124,
|
| 456 |
+
"eval_samples_per_second": 474.068,
|
| 457 |
+
"eval_steps_per_second": 7.934,
|
| 458 |
+
"eval_wer": 0.09651317984651318,
|
| 459 |
+
"step": 3000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 29.25,
|
| 463 |
+
"learning_rate": 1.7890233943843542e-05,
|
| 464 |
+
"loss": 0.2905,
|
| 465 |
+
"step": 3100
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 29.25,
|
| 469 |
+
"eval_loss": 0.4305020570755005,
|
| 470 |
+
"eval_runtime": 1.7063,
|
| 471 |
+
"eval_samples_per_second": 420.202,
|
| 472 |
+
"eval_steps_per_second": 7.033,
|
| 473 |
+
"eval_wer": 0.09109109109109109,
|
| 474 |
+
"step": 3100
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 30.19,
|
| 478 |
+
"learning_rate": 1.7683678340739462e-05,
|
| 479 |
+
"loss": 0.2964,
|
| 480 |
+
"step": 3200
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 30.19,
|
| 484 |
+
"eval_loss": 0.4379400908946991,
|
| 485 |
+
"eval_runtime": 1.587,
|
| 486 |
+
"eval_samples_per_second": 451.796,
|
| 487 |
+
"eval_steps_per_second": 7.561,
|
| 488 |
+
"eval_wer": 0.09901568234901569,
|
| 489 |
+
"step": 3200
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 31.13,
|
| 493 |
+
"learning_rate": 1.7468791051052658e-05,
|
| 494 |
+
"loss": 0.302,
|
| 495 |
+
"step": 3300
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 31.13,
|
| 499 |
+
"eval_loss": 0.4378632605075836,
|
| 500 |
+
"eval_runtime": 1.53,
|
| 501 |
+
"eval_samples_per_second": 468.64,
|
| 502 |
+
"eval_steps_per_second": 7.843,
|
| 503 |
+
"eval_wer": 0.09434434434434434,
|
| 504 |
+
"step": 3300
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 32.08,
|
| 508 |
+
"learning_rate": 1.724580508473864e-05,
|
| 509 |
+
"loss": 0.2576,
|
| 510 |
+
"step": 3400
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 32.08,
|
| 514 |
+
"eval_loss": 0.4293440580368042,
|
| 515 |
+
"eval_runtime": 1.5058,
|
| 516 |
+
"eval_samples_per_second": 476.172,
|
| 517 |
+
"eval_steps_per_second": 7.969,
|
| 518 |
+
"eval_wer": 0.09325992659325992,
|
| 519 |
+
"step": 3400
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 33.02,
|
| 523 |
+
"learning_rate": 1.701496223343694e-05,
|
| 524 |
+
"loss": 0.2771,
|
| 525 |
+
"step": 3500
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 33.02,
|
| 529 |
+
"eval_loss": 0.4239462912082672,
|
| 530 |
+
"eval_runtime": 1.6211,
|
| 531 |
+
"eval_samples_per_second": 442.284,
|
| 532 |
+
"eval_steps_per_second": 7.402,
|
| 533 |
+
"eval_wer": 0.09284284284284285,
|
| 534 |
+
"step": 3500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 33.96,
|
| 538 |
+
"learning_rate": 1.6776512808287824e-05,
|
| 539 |
+
"loss": 0.268,
|
| 540 |
+
"step": 3600
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 33.96,
|
| 544 |
+
"eval_loss": 0.4228218197822571,
|
| 545 |
+
"eval_runtime": 1.5118,
|
| 546 |
+
"eval_samples_per_second": 474.255,
|
| 547 |
+
"eval_steps_per_second": 7.937,
|
| 548 |
+
"eval_wer": 0.08942275608942275,
|
| 549 |
+
"step": 3600
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 34.91,
|
| 553 |
+
"learning_rate": 1.653071536851099e-05,
|
| 554 |
+
"loss": 0.2458,
|
| 555 |
+
"step": 3700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 34.91,
|
| 559 |
+
"eval_loss": 0.42878738045692444,
|
| 560 |
+
"eval_runtime": 1.6354,
|
| 561 |
+
"eval_samples_per_second": 438.42,
|
| 562 |
+
"eval_steps_per_second": 7.338,
|
| 563 |
+
"eval_wer": 0.08992325658992326,
|
| 564 |
+
"step": 3700
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 35.85,
|
| 568 |
+
"learning_rate": 1.627783644104057e-05,
|
| 569 |
+
"loss": 0.2553,
|
| 570 |
+
"step": 3800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 35.85,
|
| 574 |
+
"eval_loss": 0.4312421977519989,
|
| 575 |
+
"eval_runtime": 1.6977,
|
| 576 |
+
"eval_samples_per_second": 422.341,
|
| 577 |
+
"eval_steps_per_second": 7.068,
|
| 578 |
+
"eval_wer": 0.0965965965965966,
|
| 579 |
+
"step": 3800
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 36.79,
|
| 583 |
+
"learning_rate": 1.6018150231520486e-05,
|
| 584 |
+
"loss": 0.2424,
|
| 585 |
+
"step": 3900
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 36.79,
|
| 589 |
+
"eval_loss": 0.4162088632583618,
|
| 590 |
+
"eval_runtime": 1.5323,
|
| 591 |
+
"eval_samples_per_second": 467.924,
|
| 592 |
+
"eval_steps_per_second": 7.831,
|
| 593 |
+
"eval_wer": 0.09167500834167501,
|
| 594 |
+
"step": 3900
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 37.74,
|
| 598 |
+
"learning_rate": 1.575193832697346e-05,
|
| 599 |
+
"loss": 0.2501,
|
| 600 |
+
"step": 4000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 37.74,
|
| 604 |
+
"eval_loss": 0.4087802469730377,
|
| 605 |
+
"eval_runtime": 1.6438,
|
| 606 |
+
"eval_samples_per_second": 436.174,
|
| 607 |
+
"eval_steps_per_second": 7.3,
|
| 608 |
+
"eval_wer": 0.08400066733400066,
|
| 609 |
+
"step": 4000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 38.68,
|
| 613 |
+
"learning_rate": 1.5479489390466186e-05,
|
| 614 |
+
"loss": 0.2498,
|
| 615 |
+
"step": 4100
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 38.68,
|
| 619 |
+
"eval_loss": 0.4144260585308075,
|
| 620 |
+
"eval_runtime": 1.6935,
|
| 621 |
+
"eval_samples_per_second": 423.382,
|
| 622 |
+
"eval_steps_per_second": 7.086,
|
| 623 |
+
"eval_wer": 0.0920920920920921,
|
| 624 |
+
"step": 4100
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 39.62,
|
| 628 |
+
"learning_rate": 1.520109884810164e-05,
|
| 629 |
+
"loss": 0.2273,
|
| 630 |
+
"step": 4200
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 39.62,
|
| 634 |
+
"eval_loss": 0.415446400642395,
|
| 635 |
+
"eval_runtime": 1.5219,
|
| 636 |
+
"eval_samples_per_second": 471.115,
|
| 637 |
+
"eval_steps_per_second": 7.885,
|
| 638 |
+
"eval_wer": 0.08633633633633633,
|
| 639 |
+
"step": 4200
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 40.57,
|
| 643 |
+
"learning_rate": 1.4917068568678037e-05,
|
| 644 |
+
"loss": 0.23,
|
| 645 |
+
"step": 4300
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 40.57,
|
| 649 |
+
"eval_loss": 0.4157298505306244,
|
| 650 |
+
"eval_runtime": 1.689,
|
| 651 |
+
"eval_samples_per_second": 424.522,
|
| 652 |
+
"eval_steps_per_second": 7.105,
|
| 653 |
+
"eval_wer": 0.08675342008675342,
|
| 654 |
+
"step": 4300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 41.51,
|
| 658 |
+
"learning_rate": 1.4627706536361717e-05,
|
| 659 |
+
"loss": 0.2409,
|
| 660 |
+
"step": 4400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 41.51,
|
| 664 |
+
"eval_loss": 0.4032723009586334,
|
| 665 |
+
"eval_runtime": 1.5687,
|
| 666 |
+
"eval_samples_per_second": 457.076,
|
| 667 |
+
"eval_steps_per_second": 7.65,
|
| 668 |
+
"eval_wer": 0.08258258258258258,
|
| 669 |
+
"step": 4400
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 42.45,
|
| 673 |
+
"learning_rate": 1.4333326516728909e-05,
|
| 674 |
+
"loss": 0.248,
|
| 675 |
+
"step": 4500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 42.45,
|
| 679 |
+
"eval_loss": 0.4121722877025604,
|
| 680 |
+
"eval_runtime": 1.6325,
|
| 681 |
+
"eval_samples_per_second": 439.195,
|
| 682 |
+
"eval_steps_per_second": 7.351,
|
| 683 |
+
"eval_wer": 0.08466800133466801,
|
| 684 |
+
"step": 4500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 43.4,
|
| 688 |
+
"learning_rate": 1.4034247716538545e-05,
|
| 689 |
+
"loss": 0.218,
|
| 690 |
+
"step": 4600
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 43.4,
|
| 694 |
+
"eval_loss": 0.40519294142723083,
|
| 695 |
+
"eval_runtime": 1.5316,
|
| 696 |
+
"eval_samples_per_second": 468.141,
|
| 697 |
+
"eval_steps_per_second": 7.835,
|
| 698 |
+
"eval_wer": 0.08475141808475142,
|
| 699 |
+
"step": 4600
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 44.34,
|
| 703 |
+
"learning_rate": 1.373079443760497e-05,
|
| 704 |
+
"loss": 0.1979,
|
| 705 |
+
"step": 4700
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 44.34,
|
| 709 |
+
"eval_loss": 0.4063194692134857,
|
| 710 |
+
"eval_runtime": 1.6002,
|
| 711 |
+
"eval_samples_per_second": 448.081,
|
| 712 |
+
"eval_steps_per_second": 7.499,
|
| 713 |
+
"eval_wer": 0.088672005338672,
|
| 714 |
+
"step": 4700
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 45.28,
|
| 718 |
+
"learning_rate": 1.3423295725145908e-05,
|
| 719 |
+
"loss": 0.2091,
|
| 720 |
+
"step": 4800
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 45.28,
|
| 724 |
+
"eval_loss": 0.40777409076690674,
|
| 725 |
+
"eval_runtime": 1.5155,
|
| 726 |
+
"eval_samples_per_second": 473.123,
|
| 727 |
+
"eval_steps_per_second": 7.918,
|
| 728 |
+
"eval_wer": 0.08233233233233234,
|
| 729 |
+
"step": 4800
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 46.23,
|
| 733 |
+
"learning_rate": 1.3112085010987002e-05,
|
| 734 |
+
"loss": 0.2097,
|
| 735 |
+
"step": 4900
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 46.23,
|
| 739 |
+
"eval_loss": 0.4176665246486664,
|
| 740 |
+
"eval_runtime": 1.5191,
|
| 741 |
+
"eval_samples_per_second": 471.983,
|
| 742 |
+
"eval_steps_per_second": 7.899,
|
| 743 |
+
"eval_wer": 0.08933933933933934,
|
| 744 |
+
"step": 4900
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 47.17,
|
| 748 |
+
"learning_rate": 1.279749975200978e-05,
|
| 749 |
+
"loss": 0.2017,
|
| 750 |
+
"step": 5000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 47.17,
|
| 754 |
+
"eval_loss": 0.4294622540473938,
|
| 755 |
+
"eval_runtime": 1.6751,
|
| 756 |
+
"eval_samples_per_second": 428.045,
|
| 757 |
+
"eval_steps_per_second": 7.164,
|
| 758 |
+
"eval_wer": 0.088672005338672,
|
| 759 |
+
"step": 5000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 48.11,
|
| 763 |
+
"learning_rate": 1.247988106423513e-05,
|
| 764 |
+
"loss": 0.1899,
|
| 765 |
+
"step": 5100
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 48.11,
|
| 769 |
+
"eval_loss": 0.4177272915840149,
|
| 770 |
+
"eval_runtime": 1.5405,
|
| 771 |
+
"eval_samples_per_second": 465.436,
|
| 772 |
+
"eval_steps_per_second": 7.79,
|
| 773 |
+
"eval_wer": 0.09192525859192525,
|
| 774 |
+
"step": 5100
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 49.06,
|
| 778 |
+
"learning_rate": 1.2159573352939014e-05,
|
| 779 |
+
"loss": 0.195,
|
| 780 |
+
"step": 5200
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 49.06,
|
| 784 |
+
"eval_loss": 0.41085565090179443,
|
| 785 |
+
"eval_runtime": 1.5321,
|
| 786 |
+
"eval_samples_per_second": 467.995,
|
| 787 |
+
"eval_steps_per_second": 7.833,
|
| 788 |
+
"eval_wer": 0.08800467133800467,
|
| 789 |
+
"step": 5200
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 50.0,
|
| 793 |
+
"learning_rate": 1.1836923939201543e-05,
|
| 794 |
+
"loss": 0.179,
|
| 795 |
+
"step": 5300
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 50.0,
|
| 799 |
+
"eval_loss": 0.4089372456073761,
|
| 800 |
+
"eval_runtime": 1.5066,
|
| 801 |
+
"eval_samples_per_second": 475.915,
|
| 802 |
+
"eval_steps_per_second": 7.965,
|
| 803 |
+
"eval_wer": 0.08792125458792126,
|
| 804 |
+
"step": 5300
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 50.94,
|
| 808 |
+
"learning_rate": 1.1512282683294342e-05,
|
| 809 |
+
"loss": 0.1773,
|
| 810 |
+
"step": 5400
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 50.94,
|
| 814 |
+
"eval_loss": 0.40705129504203796,
|
| 815 |
+
"eval_runtime": 1.5026,
|
| 816 |
+
"eval_samples_per_second": 477.179,
|
| 817 |
+
"eval_steps_per_second": 7.986,
|
| 818 |
+
"eval_wer": 0.08425091758425092,
|
| 819 |
+
"step": 5400
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 51.89,
|
| 823 |
+
"learning_rate": 1.1186001605314571e-05,
|
| 824 |
+
"loss": 0.1889,
|
| 825 |
+
"step": 5500
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 51.89,
|
| 829 |
+
"eval_loss": 0.40717387199401855,
|
| 830 |
+
"eval_runtime": 1.6838,
|
| 831 |
+
"eval_samples_per_second": 425.834,
|
| 832 |
+
"eval_steps_per_second": 7.127,
|
| 833 |
+
"eval_wer": 0.08850517183850518,
|
| 834 |
+
"step": 5500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 52.83,
|
| 838 |
+
"learning_rate": 1.0858434503476969e-05,
|
| 839 |
+
"loss": 0.1987,
|
| 840 |
+
"step": 5600
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 52.83,
|
| 844 |
+
"eval_loss": 0.403276652097702,
|
| 845 |
+
"eval_runtime": 1.6752,
|
| 846 |
+
"eval_samples_per_second": 428.011,
|
| 847 |
+
"eval_steps_per_second": 7.163,
|
| 848 |
+
"eval_wer": 0.08733733733733734,
|
| 849 |
+
"step": 5600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 53.77,
|
| 853 |
+
"learning_rate": 1.052993657047783e-05,
|
| 854 |
+
"loss": 0.1979,
|
| 855 |
+
"step": 5700
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 53.77,
|
| 859 |
+
"eval_loss": 0.403276652097702,
|
| 860 |
+
"eval_runtime": 1.5889,
|
| 861 |
+
"eval_samples_per_second": 451.268,
|
| 862 |
+
"eval_steps_per_second": 7.553,
|
| 863 |
+
"eval_wer": 0.09275942609275943,
|
| 864 |
+
"step": 5700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 54.72,
|
| 868 |
+
"learning_rate": 1.02008640083469e-05,
|
| 869 |
+
"loss": 0.1777,
|
| 870 |
+
"step": 5800
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 54.72,
|
| 874 |
+
"eval_loss": 0.4077288806438446,
|
| 875 |
+
"eval_runtime": 1.4993,
|
| 876 |
+
"eval_samples_per_second": 478.216,
|
| 877 |
+
"eval_steps_per_second": 8.004,
|
| 878 |
+
"eval_wer": 0.08975642308975643,
|
| 879 |
+
"step": 5800
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 55.66,
|
| 883 |
+
"learning_rate": 9.871573642204777e-06,
|
| 884 |
+
"loss": 0.1742,
|
| 885 |
+
"step": 5900
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 55.66,
|
| 889 |
+
"eval_loss": 0.3968764543533325,
|
| 890 |
+
"eval_runtime": 1.6766,
|
| 891 |
+
"eval_samples_per_second": 427.66,
|
| 892 |
+
"eval_steps_per_second": 7.157,
|
| 893 |
+
"eval_wer": 0.08383383383383383,
|
| 894 |
+
"step": 5900
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 56.6,
|
| 898 |
+
"learning_rate": 9.545712182877714e-06,
|
| 899 |
+
"loss": 0.1678,
|
| 900 |
+
"step": 6000
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 56.6,
|
| 904 |
+
"eval_loss": 0.39965710043907166,
|
| 905 |
+
"eval_runtime": 1.5536,
|
| 906 |
+
"eval_samples_per_second": 461.501,
|
| 907 |
+
"eval_steps_per_second": 7.724,
|
| 908 |
+
"eval_wer": 0.08058058058058058,
|
| 909 |
+
"step": 6000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 57.55,
|
| 913 |
+
"learning_rate": 9.21705051465562e-06,
|
| 914 |
+
"loss": 0.1726,
|
| 915 |
+
"step": 6100
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 57.55,
|
| 919 |
+
"eval_loss": 0.3978481888771057,
|
| 920 |
+
"eval_runtime": 1.5144,
|
| 921 |
+
"eval_samples_per_second": 473.444,
|
| 922 |
+
"eval_steps_per_second": 7.924,
|
| 923 |
+
"eval_wer": 0.08850517183850518,
|
| 924 |
+
"step": 6100
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 58.49,
|
| 928 |
+
"learning_rate": 8.889237826496551e-06,
|
| 929 |
+
"loss": 0.1602,
|
| 930 |
+
"step": 6200
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 58.49,
|
| 934 |
+
"eval_loss": 0.39665135741233826,
|
| 935 |
+
"eval_runtime": 1.5277,
|
| 936 |
+
"eval_samples_per_second": 469.328,
|
| 937 |
+
"eval_steps_per_second": 7.855,
|
| 938 |
+
"eval_wer": 0.08600266933600267,
|
| 939 |
+
"step": 6200
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 59.43,
|
| 943 |
+
"learning_rate": 8.562629577394817e-06,
|
| 944 |
+
"loss": 0.1681,
|
| 945 |
+
"step": 6300
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 59.43,
|
| 949 |
+
"eval_loss": 0.4038829207420349,
|
| 950 |
+
"eval_runtime": 1.5785,
|
| 951 |
+
"eval_samples_per_second": 454.223,
|
| 952 |
+
"eval_steps_per_second": 7.602,
|
| 953 |
+
"eval_wer": 0.09009009009009009,
|
| 954 |
+
"step": 6300
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 60.38,
|
| 958 |
+
"learning_rate": 8.237579920328478e-06,
|
| 959 |
+
"loss": 0.1594,
|
| 960 |
+
"step": 6400
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 60.38,
|
| 964 |
+
"eval_loss": 0.3992190361022949,
|
| 965 |
+
"eval_runtime": 1.5157,
|
| 966 |
+
"eval_samples_per_second": 473.056,
|
| 967 |
+
"eval_steps_per_second": 7.917,
|
| 968 |
+
"eval_wer": 0.08558558558558559,
|
| 969 |
+
"step": 6400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 61.32,
|
| 973 |
+
"learning_rate": 7.91444131823864e-06,
|
| 974 |
+
"loss": 0.171,
|
| 975 |
+
"step": 6500
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 61.32,
|
| 979 |
+
"eval_loss": 0.4058407247066498,
|
| 980 |
+
"eval_runtime": 1.5604,
|
| 981 |
+
"eval_samples_per_second": 459.498,
|
| 982 |
+
"eval_steps_per_second": 7.69,
|
| 983 |
+
"eval_wer": 0.08900567233900568,
|
| 984 |
+
"step": 6500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 62.26,
|
| 988 |
+
"learning_rate": 7.593564161841318e-06,
|
| 989 |
+
"loss": 0.1691,
|
| 990 |
+
"step": 6600
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 62.26,
|
| 994 |
+
"eval_loss": 0.40775159001350403,
|
| 995 |
+
"eval_runtime": 1.6471,
|
| 996 |
+
"eval_samples_per_second": 435.301,
|
| 997 |
+
"eval_steps_per_second": 7.285,
|
| 998 |
+
"eval_wer": 0.0841675008341675,
|
| 999 |
+
"step": 6600
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 63.21,
|
| 1003 |
+
"learning_rate": 7.275296389686258e-06,
|
| 1004 |
+
"loss": 0.1724,
|
| 1005 |
+
"step": 6700
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 63.21,
|
| 1009 |
+
"eval_loss": 0.4161495268344879,
|
| 1010 |
+
"eval_runtime": 1.5379,
|
| 1011 |
+
"eval_samples_per_second": 466.219,
|
| 1012 |
+
"eval_steps_per_second": 7.803,
|
| 1013 |
+
"eval_wer": 0.09025692359025693,
|
| 1014 |
+
"step": 6700
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 64.15,
|
| 1018 |
+
"learning_rate": 6.959983110874782e-06,
|
| 1019 |
+
"loss": 0.172,
|
| 1020 |
+
"step": 6800
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 64.15,
|
| 1024 |
+
"eval_loss": 0.4121454358100891,
|
| 1025 |
+
"eval_runtime": 1.5882,
|
| 1026 |
+
"eval_samples_per_second": 451.46,
|
| 1027 |
+
"eval_steps_per_second": 7.556,
|
| 1028 |
+
"eval_wer": 0.08992325658992326,
|
| 1029 |
+
"step": 6800
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 65.09,
|
| 1033 |
+
"learning_rate": 6.647966230845655e-06,
|
| 1034 |
+
"loss": 0.1717,
|
| 1035 |
+
"step": 6900
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 65.09,
|
| 1039 |
+
"eval_loss": 0.4110967218875885,
|
| 1040 |
+
"eval_runtime": 1.6285,
|
| 1041 |
+
"eval_samples_per_second": 440.291,
|
| 1042 |
+
"eval_steps_per_second": 7.369,
|
| 1043 |
+
"eval_wer": 0.08783783783783784,
|
| 1044 |
+
"step": 6900
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 66.04,
|
| 1048 |
+
"learning_rate": 6.339584080634824e-06,
|
| 1049 |
+
"loss": 0.1775,
|
| 1050 |
+
"step": 7000
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 66.04,
|
| 1054 |
+
"eval_loss": 0.4108857810497284,
|
| 1055 |
+
"eval_runtime": 1.6369,
|
| 1056 |
+
"eval_samples_per_second": 438.022,
|
| 1057 |
+
"eval_steps_per_second": 7.331,
|
| 1058 |
+
"eval_wer": 0.09259259259259259,
|
| 1059 |
+
"step": 7000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 66.98,
|
| 1063 |
+
"learning_rate": 6.03517105001101e-06,
|
| 1064 |
+
"loss": 0.1607,
|
| 1065 |
+
"step": 7100
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 66.98,
|
| 1069 |
+
"eval_loss": 0.40803730487823486,
|
| 1070 |
+
"eval_runtime": 1.5893,
|
| 1071 |
+
"eval_samples_per_second": 451.133,
|
| 1072 |
+
"eval_steps_per_second": 7.55,
|
| 1073 |
+
"eval_wer": 0.09075742409075742,
|
| 1074 |
+
"step": 7100
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 67.92,
|
| 1078 |
+
"learning_rate": 5.735057224884939e-06,
|
| 1079 |
+
"loss": 0.1606,
|
| 1080 |
+
"step": 7200
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 67.92,
|
| 1084 |
+
"eval_loss": 0.4069909453392029,
|
| 1085 |
+
"eval_runtime": 1.52,
|
| 1086 |
+
"eval_samples_per_second": 471.701,
|
| 1087 |
+
"eval_steps_per_second": 7.895,
|
| 1088 |
+
"eval_wer": 0.09300967634300968,
|
| 1089 |
+
"step": 7200
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 68.87,
|
| 1093 |
+
"learning_rate": 5.439568029385422e-06,
|
| 1094 |
+
"loss": 0.1801,
|
| 1095 |
+
"step": 7300
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 68.87,
|
| 1099 |
+
"eval_loss": 0.4096461832523346,
|
| 1100 |
+
"eval_runtime": 1.5795,
|
| 1101 |
+
"eval_samples_per_second": 453.931,
|
| 1102 |
+
"eval_steps_per_second": 7.597,
|
| 1103 |
+
"eval_wer": 0.09075742409075742,
|
| 1104 |
+
"step": 7300
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 69.81,
|
| 1108 |
+
"learning_rate": 5.14902387299036e-06,
|
| 1109 |
+
"loss": 0.16,
|
| 1110 |
+
"step": 7400
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 69.81,
|
| 1114 |
+
"eval_loss": 0.4029737114906311,
|
| 1115 |
+
"eval_runtime": 1.6293,
|
| 1116 |
+
"eval_samples_per_second": 440.075,
|
| 1117 |
+
"eval_steps_per_second": 7.365,
|
| 1118 |
+
"eval_wer": 0.09325992659325992,
|
| 1119 |
+
"step": 7400
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 70.75,
|
| 1123 |
+
"learning_rate": 4.863739803095299e-06,
|
| 1124 |
+
"loss": 0.1433,
|
| 1125 |
+
"step": 7500
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 70.75,
|
| 1129 |
+
"eval_loss": 0.40587371587753296,
|
| 1130 |
+
"eval_runtime": 1.5048,
|
| 1131 |
+
"eval_samples_per_second": 476.48,
|
| 1132 |
+
"eval_steps_per_second": 7.975,
|
| 1133 |
+
"eval_wer": 0.09200867534200867,
|
| 1134 |
+
"step": 7500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 71.7,
|
| 1138 |
+
"learning_rate": 4.584025163396323e-06,
|
| 1139 |
+
"loss": 0.1473,
|
| 1140 |
+
"step": 7600
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 71.7,
|
| 1144 |
+
"eval_loss": 0.4119517207145691,
|
| 1145 |
+
"eval_runtime": 1.5044,
|
| 1146 |
+
"eval_samples_per_second": 476.615,
|
| 1147 |
+
"eval_steps_per_second": 7.977,
|
| 1148 |
+
"eval_wer": 0.09793126459793126,
|
| 1149 |
+
"step": 7600
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 72.64,
|
| 1153 |
+
"learning_rate": 4.310183258457632e-06,
|
| 1154 |
+
"loss": 0.1396,
|
| 1155 |
+
"step": 7700
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 72.64,
|
| 1159 |
+
"eval_loss": 0.4061850905418396,
|
| 1160 |
+
"eval_runtime": 1.5615,
|
| 1161 |
+
"eval_samples_per_second": 459.166,
|
| 1162 |
+
"eval_steps_per_second": 7.685,
|
| 1163 |
+
"eval_wer": 0.09217550884217551,
|
| 1164 |
+
"step": 7700
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 73.58,
|
| 1168 |
+
"learning_rate": 4.0451562485624955e-06,
|
| 1169 |
+
"loss": 0.1429,
|
| 1170 |
+
"step": 7800
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 73.58,
|
| 1174 |
+
"eval_loss": 0.40794309973716736,
|
| 1175 |
+
"eval_runtime": 1.6245,
|
| 1176 |
+
"eval_samples_per_second": 441.376,
|
| 1177 |
+
"eval_steps_per_second": 7.387,
|
| 1178 |
+
"eval_wer": 0.08992325658992326,
|
| 1179 |
+
"step": 7800
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 74.53,
|
| 1183 |
+
"learning_rate": 3.7838779196384036e-06,
|
| 1184 |
+
"loss": 0.1332,
|
| 1185 |
+
"step": 7900
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 74.53,
|
| 1189 |
+
"eval_loss": 0.4055492877960205,
|
| 1190 |
+
"eval_runtime": 1.5063,
|
| 1191 |
+
"eval_samples_per_second": 475.996,
|
| 1192 |
+
"eval_steps_per_second": 7.966,
|
| 1193 |
+
"eval_wer": 0.08508508508508508,
|
| 1194 |
+
"step": 7900
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 75.47,
|
| 1198 |
+
"learning_rate": 3.5293399536806903e-06,
|
| 1199 |
+
"loss": 0.1429,
|
| 1200 |
+
"step": 8000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 75.47,
|
| 1204 |
+
"eval_loss": 0.40807628631591797,
|
| 1205 |
+
"eval_runtime": 1.5786,
|
| 1206 |
+
"eval_samples_per_second": 454.202,
|
| 1207 |
+
"eval_steps_per_second": 7.602,
|
| 1208 |
+
"eval_wer": 0.09217550884217551,
|
| 1209 |
+
"step": 8000
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 76.42,
|
| 1213 |
+
"learning_rate": 3.2818183552879e-06,
|
| 1214 |
+
"loss": 0.1528,
|
| 1215 |
+
"step": 8100
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 76.42,
|
| 1219 |
+
"eval_loss": 0.40834707021713257,
|
| 1220 |
+
"eval_runtime": 1.516,
|
| 1221 |
+
"eval_samples_per_second": 472.942,
|
| 1222 |
+
"eval_steps_per_second": 7.915,
|
| 1223 |
+
"eval_wer": 0.08533533533533534,
|
| 1224 |
+
"step": 8100
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 77.36,
|
| 1228 |
+
"learning_rate": 3.041581520960881e-06,
|
| 1229 |
+
"loss": 0.1547,
|
| 1230 |
+
"step": 8200
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 77.36,
|
| 1234 |
+
"eval_loss": 0.41393765807151794,
|
| 1235 |
+
"eval_runtime": 1.5089,
|
| 1236 |
+
"eval_samples_per_second": 475.178,
|
| 1237 |
+
"eval_steps_per_second": 7.953,
|
| 1238 |
+
"eval_wer": 0.09451117784451117,
|
| 1239 |
+
"step": 8200
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 78.3,
|
| 1243 |
+
"learning_rate": 2.8088899480708677e-06,
|
| 1244 |
+
"loss": 0.1384,
|
| 1245 |
+
"step": 8300
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 78.3,
|
| 1249 |
+
"eval_loss": 0.4110879600048065,
|
| 1250 |
+
"eval_runtime": 1.5103,
|
| 1251 |
+
"eval_samples_per_second": 474.75,
|
| 1252 |
+
"eval_steps_per_second": 7.946,
|
| 1253 |
+
"eval_wer": 0.09325992659325992,
|
| 1254 |
+
"step": 8300
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 79.25,
|
| 1258 |
+
"learning_rate": 2.5839959523929126e-06,
|
| 1259 |
+
"loss": 0.1696,
|
| 1260 |
+
"step": 8400
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 79.25,
|
| 1264 |
+
"eval_loss": 0.41319677233695984,
|
| 1265 |
+
"eval_runtime": 1.6749,
|
| 1266 |
+
"eval_samples_per_second": 428.087,
|
| 1267 |
+
"eval_steps_per_second": 7.165,
|
| 1268 |
+
"eval_wer": 0.09434434434434434,
|
| 1269 |
+
"step": 8400
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 80.19,
|
| 1273 |
+
"learning_rate": 2.3671433945108712e-06,
|
| 1274 |
+
"loss": 0.1483,
|
| 1275 |
+
"step": 8500
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 80.19,
|
| 1279 |
+
"eval_loss": 0.4139257073402405,
|
| 1280 |
+
"eval_runtime": 1.5694,
|
| 1281 |
+
"eval_samples_per_second": 456.87,
|
| 1282 |
+
"eval_steps_per_second": 7.646,
|
| 1283 |
+
"eval_wer": 0.09059059059059059,
|
| 1284 |
+
"step": 8500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 81.13,
|
| 1288 |
+
"learning_rate": 2.15856741539067e-06,
|
| 1289 |
+
"loss": 0.1547,
|
| 1290 |
+
"step": 8600
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 81.13,
|
| 1294 |
+
"eval_loss": 0.4156300127506256,
|
| 1295 |
+
"eval_runtime": 1.6731,
|
| 1296 |
+
"eval_samples_per_second": 428.558,
|
| 1297 |
+
"eval_steps_per_second": 7.173,
|
| 1298 |
+
"eval_wer": 0.09592926259592927,
|
| 1299 |
+
"step": 8600
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 82.08,
|
| 1303 |
+
"learning_rate": 1.9584941814085756e-06,
|
| 1304 |
+
"loss": 0.149,
|
| 1305 |
+
"step": 8700
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 82.08,
|
| 1309 |
+
"eval_loss": 0.411948025226593,
|
| 1310 |
+
"eval_runtime": 1.5259,
|
| 1311 |
+
"eval_samples_per_second": 469.893,
|
| 1312 |
+
"eval_steps_per_second": 7.864,
|
| 1313 |
+
"eval_wer": 0.09050717384050717,
|
| 1314 |
+
"step": 8700
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 83.02,
|
| 1318 |
+
"learning_rate": 1.7671406391109204e-06,
|
| 1319 |
+
"loss": 0.1294,
|
| 1320 |
+
"step": 8800
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 83.02,
|
| 1324 |
+
"eval_loss": 0.41449442505836487,
|
| 1325 |
+
"eval_runtime": 1.6397,
|
| 1326 |
+
"eval_samples_per_second": 437.287,
|
| 1327 |
+
"eval_steps_per_second": 7.319,
|
| 1328 |
+
"eval_wer": 0.09451117784451117,
|
| 1329 |
+
"step": 8800
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 83.96,
|
| 1333 |
+
"learning_rate": 1.5847142799712112e-06,
|
| 1334 |
+
"loss": 0.1383,
|
| 1335 |
+
"step": 8900
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 83.96,
|
| 1339 |
+
"eval_loss": 0.415103554725647,
|
| 1340 |
+
"eval_runtime": 1.5828,
|
| 1341 |
+
"eval_samples_per_second": 452.985,
|
| 1342 |
+
"eval_steps_per_second": 7.581,
|
| 1343 |
+
"eval_wer": 0.09167500834167501,
|
| 1344 |
+
"step": 8900
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 84.91,
|
| 1348 |
+
"learning_rate": 1.4114129153997401e-06,
|
| 1349 |
+
"loss": 0.1356,
|
| 1350 |
+
"step": 9000
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 84.91,
|
| 1354 |
+
"eval_loss": 0.4164859354496002,
|
| 1355 |
+
"eval_runtime": 1.5707,
|
| 1356 |
+
"eval_samples_per_second": 456.48,
|
| 1357 |
+
"eval_steps_per_second": 7.64,
|
| 1358 |
+
"eval_wer": 0.09517851184517852,
|
| 1359 |
+
"step": 9000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 85.85,
|
| 1363 |
+
"learning_rate": 1.2490176555326306e-06,
|
| 1364 |
+
"loss": 0.1491,
|
| 1365 |
+
"step": 9100
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 85.85,
|
| 1369 |
+
"eval_loss": 0.41879215836524963,
|
| 1370 |
+
"eval_runtime": 1.681,
|
| 1371 |
+
"eval_samples_per_second": 426.538,
|
| 1372 |
+
"eval_steps_per_second": 7.139,
|
| 1373 |
+
"eval_wer": 0.09501167834501167,
|
| 1374 |
+
"step": 9100
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 86.79,
|
| 1378 |
+
"learning_rate": 1.0944241784701025e-06,
|
| 1379 |
+
"loss": 0.1395,
|
| 1380 |
+
"step": 9200
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 86.79,
|
| 1384 |
+
"eval_loss": 0.4173920154571533,
|
| 1385 |
+
"eval_runtime": 1.568,
|
| 1386 |
+
"eval_samples_per_second": 457.261,
|
| 1387 |
+
"eval_steps_per_second": 7.653,
|
| 1388 |
+
"eval_wer": 0.09501167834501167,
|
| 1389 |
+
"step": 9200
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 87.74,
|
| 1393 |
+
"learning_rate": 9.508883783214695e-07,
|
| 1394 |
+
"loss": 0.1439,
|
| 1395 |
+
"step": 9300
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 87.74,
|
| 1399 |
+
"eval_loss": 0.41514286398887634,
|
| 1400 |
+
"eval_runtime": 1.6196,
|
| 1401 |
+
"eval_samples_per_second": 442.706,
|
| 1402 |
+
"eval_steps_per_second": 7.409,
|
| 1403 |
+
"eval_wer": 0.09192525859192525,
|
| 1404 |
+
"step": 9300
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 88.68,
|
| 1408 |
+
"learning_rate": 8.156664476124932e-07,
|
| 1409 |
+
"loss": 0.1421,
|
| 1410 |
+
"step": 9400
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 88.68,
|
| 1414 |
+
"eval_loss": 0.4151535630226135,
|
| 1415 |
+
"eval_runtime": 1.5673,
|
| 1416 |
+
"eval_samples_per_second": 457.489,
|
| 1417 |
+
"eval_steps_per_second": 7.657,
|
| 1418 |
+
"eval_wer": 0.09309309309309309,
|
| 1419 |
+
"step": 9400
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 89.62,
|
| 1423 |
+
"learning_rate": 6.904034174764318e-07,
|
| 1424 |
+
"loss": 0.1443,
|
| 1425 |
+
"step": 9500
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 89.62,
|
| 1429 |
+
"eval_loss": 0.4160225987434387,
|
| 1430 |
+
"eval_runtime": 1.6255,
|
| 1431 |
+
"eval_samples_per_second": 441.086,
|
| 1432 |
+
"eval_steps_per_second": 7.382,
|
| 1433 |
+
"eval_wer": 0.09442776109442776,
|
| 1434 |
+
"step": 9500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 90.57,
|
| 1438 |
+
"learning_rate": 5.752351150863133e-07,
|
| 1439 |
+
"loss": 0.1429,
|
| 1440 |
+
"step": 9600
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 90.57,
|
| 1444 |
+
"eval_loss": 0.41378459334373474,
|
| 1445 |
+
"eval_runtime": 1.6157,
|
| 1446 |
+
"eval_samples_per_second": 443.764,
|
| 1447 |
+
"eval_steps_per_second": 7.427,
|
| 1448 |
+
"eval_wer": 0.09284284284284285,
|
| 1449 |
+
"step": 9600
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 91.51,
|
| 1453 |
+
"learning_rate": 4.702864215416847e-07,
|
| 1454 |
+
"loss": 0.1397,
|
| 1455 |
+
"step": 9700
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 91.51,
|
| 1459 |
+
"eval_loss": 0.4148540794849396,
|
| 1460 |
+
"eval_runtime": 1.499,
|
| 1461 |
+
"eval_samples_per_second": 478.314,
|
| 1462 |
+
"eval_steps_per_second": 8.005,
|
| 1463 |
+
"eval_wer": 0.09184184184184184,
|
| 1464 |
+
"step": 9700
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 92.45,
|
| 1468 |
+
"learning_rate": 3.7567113645558537e-07,
|
| 1469 |
+
"loss": 0.155,
|
| 1470 |
+
"step": 9800
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 92.45,
|
| 1474 |
+
"eval_loss": 0.4144271910190582,
|
| 1475 |
+
"eval_runtime": 1.6208,
|
| 1476 |
+
"eval_samples_per_second": 442.371,
|
| 1477 |
+
"eval_steps_per_second": 7.404,
|
| 1478 |
+
"eval_wer": 0.09150817484150818,
|
| 1479 |
+
"step": 9800
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 93.4,
|
| 1483 |
+
"learning_rate": 2.914918545575618e-07,
|
| 1484 |
+
"loss": 0.1406,
|
| 1485 |
+
"step": 9900
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 93.4,
|
| 1489 |
+
"eval_loss": 0.41388100385665894,
|
| 1490 |
+
"eval_runtime": 1.5267,
|
| 1491 |
+
"eval_samples_per_second": 469.648,
|
| 1492 |
+
"eval_steps_per_second": 7.86,
|
| 1493 |
+
"eval_wer": 0.0920920920920921,
|
| 1494 |
+
"step": 9900
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 94.34,
|
| 1498 |
+
"learning_rate": 2.1852399266194312e-07,
|
| 1499 |
+
"loss": 0.1328,
|
| 1500 |
+
"step": 10000
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 94.34,
|
| 1504 |
+
"eval_loss": 0.414026141166687,
|
| 1505 |
+
"eval_runtime": 1.566,
|
| 1506 |
+
"eval_samples_per_second": 457.869,
|
| 1507 |
+
"eval_steps_per_second": 7.663,
|
| 1508 |
+
"eval_wer": 0.09292625959292626,
|
| 1509 |
+
"step": 10000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 95.28,
|
| 1513 |
+
"learning_rate": 1.5537270875273348e-07,
|
| 1514 |
+
"loss": 0.1461,
|
| 1515 |
+
"step": 10100
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 95.28,
|
| 1519 |
+
"eval_loss": 0.41416749358177185,
|
| 1520 |
+
"eval_runtime": 1.6735,
|
| 1521 |
+
"eval_samples_per_second": 428.446,
|
| 1522 |
+
"eval_steps_per_second": 7.171,
|
| 1523 |
+
"eval_wer": 0.09142475809142475,
|
| 1524 |
+
"step": 10100
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 96.23,
|
| 1528 |
+
"learning_rate": 1.028963054771226e-07,
|
| 1529 |
+
"loss": 0.1455,
|
| 1530 |
+
"step": 10200
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 96.23,
|
| 1534 |
+
"eval_loss": 0.41422799229621887,
|
| 1535 |
+
"eval_runtime": 1.6192,
|
| 1536 |
+
"eval_samples_per_second": 442.819,
|
| 1537 |
+
"eval_steps_per_second": 7.411,
|
| 1538 |
+
"eval_wer": 0.09134134134134134,
|
| 1539 |
+
"step": 10200
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 97.17,
|
| 1543 |
+
"learning_rate": 6.115168487157097e-08,
|
| 1544 |
+
"loss": 0.155,
|
| 1545 |
+
"step": 10300
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 97.17,
|
| 1549 |
+
"eval_loss": 0.4138621985912323,
|
| 1550 |
+
"eval_runtime": 1.5571,
|
| 1551 |
+
"eval_samples_per_second": 460.473,
|
| 1552 |
+
"eval_steps_per_second": 7.707,
|
| 1553 |
+
"eval_wer": 0.09142475809142475,
|
| 1554 |
+
"step": 10300
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 98.11,
|
| 1558 |
+
"learning_rate": 3.018411211767158e-08,
|
| 1559 |
+
"loss": 0.147,
|
| 1560 |
+
"step": 10400
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 98.11,
|
| 1564 |
+
"eval_loss": 0.4140460193157196,
|
| 1565 |
+
"eval_runtime": 1.6844,
|
| 1566 |
+
"eval_samples_per_second": 425.677,
|
| 1567 |
+
"eval_steps_per_second": 7.124,
|
| 1568 |
+
"eval_wer": 0.09184184184184184,
|
| 1569 |
+
"step": 10400
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 99.06,
|
| 1573 |
+
"learning_rate": 1.002716645950197e-08,
|
| 1574 |
+
"loss": 0.1298,
|
| 1575 |
+
"step": 10500
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 99.06,
|
| 1579 |
+
"eval_loss": 0.41404303908348083,
|
| 1580 |
+
"eval_runtime": 1.6696,
|
| 1581 |
+
"eval_samples_per_second": 429.449,
|
| 1582 |
+
"eval_steps_per_second": 7.187,
|
| 1583 |
+
"eval_wer": 0.09167500834167501,
|
| 1584 |
+
"step": 10500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 100.0,
|
| 1588 |
+
"learning_rate": 7.027047924512698e-10,
|
| 1589 |
+
"loss": 0.1508,
|
| 1590 |
+
"step": 10600
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 100.0,
|
| 1594 |
+
"eval_loss": 0.41406112909317017,
|
| 1595 |
+
"eval_runtime": 1.6217,
|
| 1596 |
+
"eval_samples_per_second": 442.135,
|
| 1597 |
+
"eval_steps_per_second": 7.4,
|
| 1598 |
+
"eval_wer": 0.09142475809142475,
|
| 1599 |
+
"step": 10600
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 100.0,
|
| 1603 |
+
"step": 10600,
|
| 1604 |
+
"total_flos": 2.842046364754798e+19,
|
| 1605 |
+
"train_loss": 2.4498984623855016,
|
| 1606 |
+
"train_runtime": 3716.1555,
|
| 1607 |
+
"train_samples_per_second": 91.062,
|
| 1608 |
+
"train_steps_per_second": 2.852
|
| 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 |
+
}
|