Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +1629 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1629 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2282,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.008764241893076249,
|
| 13 |
+
"grad_norm": 3.7227299213409424,
|
| 14 |
+
"learning_rate": 9.999654593547529e-05,
|
| 15 |
+
"loss": 0.8928,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.017528483786152498,
|
| 20 |
+
"grad_norm": 2.602724552154541,
|
| 21 |
+
"learning_rate": 9.998460664009619e-05,
|
| 22 |
+
"loss": 0.7339,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.026292725679228746,
|
| 27 |
+
"grad_norm": 2.0315356254577637,
|
| 28 |
+
"learning_rate": 9.996414173332143e-05,
|
| 29 |
+
"loss": 0.8093,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.035056967572304996,
|
| 34 |
+
"grad_norm": 2.265153408050537,
|
| 35 |
+
"learning_rate": 9.993515509371871e-05,
|
| 36 |
+
"loss": 0.7516,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04382120946538125,
|
| 41 |
+
"grad_norm": 1.9205143451690674,
|
| 42 |
+
"learning_rate": 9.989765221491895e-05,
|
| 43 |
+
"loss": 0.7691,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05258545135845749,
|
| 48 |
+
"grad_norm": 2.219170331954956,
|
| 49 |
+
"learning_rate": 9.985164020457504e-05,
|
| 50 |
+
"loss": 0.7539,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06134969325153374,
|
| 55 |
+
"grad_norm": 2.3345134258270264,
|
| 56 |
+
"learning_rate": 9.979712778301489e-05,
|
| 57 |
+
"loss": 0.6932,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.07011393514460999,
|
| 62 |
+
"grad_norm": 1.9078369140625,
|
| 63 |
+
"learning_rate": 9.973412528158862e-05,
|
| 64 |
+
"loss": 0.7058,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.07887817703768624,
|
| 69 |
+
"grad_norm": 2.3402259349823,
|
| 70 |
+
"learning_rate": 9.966264464071064e-05,
|
| 71 |
+
"loss": 0.7181,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0876424189307625,
|
| 76 |
+
"grad_norm": 1.6266059875488281,
|
| 77 |
+
"learning_rate": 9.958269940759659e-05,
|
| 78 |
+
"loss": 0.6715,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.09640666082383874,
|
| 83 |
+
"grad_norm": 1.8910496234893799,
|
| 84 |
+
"learning_rate": 9.94943047336958e-05,
|
| 85 |
+
"loss": 0.6983,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.10517090271691498,
|
| 90 |
+
"grad_norm": 1.970109224319458,
|
| 91 |
+
"learning_rate": 9.939747737181993e-05,
|
| 92 |
+
"loss": 0.7639,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.11393514460999124,
|
| 97 |
+
"grad_norm": 1.9310839176177979,
|
| 98 |
+
"learning_rate": 9.929223567296766e-05,
|
| 99 |
+
"loss": 0.6289,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.12269938650306748,
|
| 104 |
+
"grad_norm": 1.5292866230010986,
|
| 105 |
+
"learning_rate": 9.917859958284699e-05,
|
| 106 |
+
"loss": 0.6575,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.13146362839614373,
|
| 111 |
+
"grad_norm": 1.3497545719146729,
|
| 112 |
+
"learning_rate": 9.905659063809492e-05,
|
| 113 |
+
"loss": 0.6986,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.14022787028921999,
|
| 118 |
+
"grad_norm": 2.1464974880218506,
|
| 119 |
+
"learning_rate": 9.892623196219586e-05,
|
| 120 |
+
"loss": 0.7055,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.14899211218229624,
|
| 125 |
+
"grad_norm": 2.5107662677764893,
|
| 126 |
+
"learning_rate": 9.878754826109915e-05,
|
| 127 |
+
"loss": 0.7421,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.15775635407537247,
|
| 132 |
+
"grad_norm": 1.9968948364257812,
|
| 133 |
+
"learning_rate": 9.864056581853674e-05,
|
| 134 |
+
"loss": 0.6956,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.16652059596844873,
|
| 139 |
+
"grad_norm": 1.5380990505218506,
|
| 140 |
+
"learning_rate": 9.84853124910418e-05,
|
| 141 |
+
"loss": 0.7439,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.175284837861525,
|
| 146 |
+
"grad_norm": 1.8293559551239014,
|
| 147 |
+
"learning_rate": 9.832181770266927e-05,
|
| 148 |
+
"loss": 0.6924,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.18404907975460122,
|
| 153 |
+
"grad_norm": 1.5421850681304932,
|
| 154 |
+
"learning_rate": 9.815011243941939e-05,
|
| 155 |
+
"loss": 0.6443,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.19281332164767748,
|
| 160 |
+
"grad_norm": 1.7461329698562622,
|
| 161 |
+
"learning_rate": 9.797022924336504e-05,
|
| 162 |
+
"loss": 0.7049,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.20157756354075373,
|
| 167 |
+
"grad_norm": 1.9567017555236816,
|
| 168 |
+
"learning_rate": 9.778220220648439e-05,
|
| 169 |
+
"loss": 0.6797,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.21034180543382996,
|
| 174 |
+
"grad_norm": 2.222736358642578,
|
| 175 |
+
"learning_rate": 9.75860669641996e-05,
|
| 176 |
+
"loss": 0.6958,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.21910604732690622,
|
| 181 |
+
"grad_norm": 1.6702378988265991,
|
| 182 |
+
"learning_rate": 9.738186068862311e-05,
|
| 183 |
+
"loss": 0.6895,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.22787028921998248,
|
| 188 |
+
"grad_norm": 1.9190706014633179,
|
| 189 |
+
"learning_rate": 9.716962208151269e-05,
|
| 190 |
+
"loss": 0.6747,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2366345311130587,
|
| 195 |
+
"grad_norm": 2.277719020843506,
|
| 196 |
+
"learning_rate": 9.69493913669366e-05,
|
| 197 |
+
"loss": 0.704,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.24539877300613497,
|
| 202 |
+
"grad_norm": 1.6245454549789429,
|
| 203 |
+
"learning_rate": 9.672121028365014e-05,
|
| 204 |
+
"loss": 0.6023,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2541630148992112,
|
| 209 |
+
"grad_norm": 1.7889143228530884,
|
| 210 |
+
"learning_rate": 9.648512207718532e-05,
|
| 211 |
+
"loss": 0.7218,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.26292725679228746,
|
| 216 |
+
"grad_norm": 2.422360897064209,
|
| 217 |
+
"learning_rate": 9.624117149165466e-05,
|
| 218 |
+
"loss": 0.6986,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.27169149868536374,
|
| 223 |
+
"grad_norm": 2.14021372795105,
|
| 224 |
+
"learning_rate": 9.598940476127131e-05,
|
| 225 |
+
"loss": 0.6476,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.28045574057843997,
|
| 230 |
+
"grad_norm": 1.9130802154541016,
|
| 231 |
+
"learning_rate": 9.57298696015866e-05,
|
| 232 |
+
"loss": 0.7056,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.2892199824715162,
|
| 237 |
+
"grad_norm": 2.4466943740844727,
|
| 238 |
+
"learning_rate": 9.546261520044675e-05,
|
| 239 |
+
"loss": 0.6542,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.2979842243645925,
|
| 244 |
+
"grad_norm": 1.6736353635787964,
|
| 245 |
+
"learning_rate": 9.518769220867076e-05,
|
| 246 |
+
"loss": 0.6648,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.3067484662576687,
|
| 251 |
+
"grad_norm": 1.745496153831482,
|
| 252 |
+
"learning_rate": 9.490515273045085e-05,
|
| 253 |
+
"loss": 0.6411,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.31551270815074495,
|
| 258 |
+
"grad_norm": 2.82519268989563,
|
| 259 |
+
"learning_rate": 9.461505031347753e-05,
|
| 260 |
+
"loss": 0.6597,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.32427695004382123,
|
| 265 |
+
"grad_norm": 2.2791597843170166,
|
| 266 |
+
"learning_rate": 9.431743993879119e-05,
|
| 267 |
+
"loss": 0.6931,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.33304119193689746,
|
| 272 |
+
"grad_norm": 1.6820751428604126,
|
| 273 |
+
"learning_rate": 9.401237801036176e-05,
|
| 274 |
+
"loss": 0.6529,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3418054338299737,
|
| 279 |
+
"grad_norm": 1.7358545064926147,
|
| 280 |
+
"learning_rate": 9.369992234439899e-05,
|
| 281 |
+
"loss": 0.7001,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.35056967572305,
|
| 286 |
+
"grad_norm": 1.7271596193313599,
|
| 287 |
+
"learning_rate": 9.338013215839495e-05,
|
| 288 |
+
"loss": 0.6801,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3593339176161262,
|
| 293 |
+
"grad_norm": 1.7952642440795898,
|
| 294 |
+
"learning_rate": 9.305306805990093e-05,
|
| 295 |
+
"loss": 0.7023,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.36809815950920244,
|
| 300 |
+
"grad_norm": 2.275848865509033,
|
| 301 |
+
"learning_rate": 9.271879203504094e-05,
|
| 302 |
+
"loss": 0.6641,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.3768624014022787,
|
| 307 |
+
"grad_norm": 1.435059666633606,
|
| 308 |
+
"learning_rate": 9.237736743676386e-05,
|
| 309 |
+
"loss": 0.6739,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.38562664329535495,
|
| 314 |
+
"grad_norm": 1.6760456562042236,
|
| 315 |
+
"learning_rate": 9.202885897283674e-05,
|
| 316 |
+
"loss": 0.6245,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.3943908851884312,
|
| 321 |
+
"grad_norm": 1.5440309047698975,
|
| 322 |
+
"learning_rate": 9.167333269358109e-05,
|
| 323 |
+
"loss": 0.6436,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.40315512708150747,
|
| 328 |
+
"grad_norm": 1.798532485961914,
|
| 329 |
+
"learning_rate": 9.131085597935487e-05,
|
| 330 |
+
"loss": 0.6097,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4119193689745837,
|
| 335 |
+
"grad_norm": 1.5643911361694336,
|
| 336 |
+
"learning_rate": 9.094149752778233e-05,
|
| 337 |
+
"loss": 0.7032,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.42068361086765993,
|
| 342 |
+
"grad_norm": 1.9498308897018433,
|
| 343 |
+
"learning_rate": 9.056532734073434e-05,
|
| 344 |
+
"loss": 0.6429,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.4294478527607362,
|
| 349 |
+
"grad_norm": 2.7362070083618164,
|
| 350 |
+
"learning_rate": 9.018241671106134e-05,
|
| 351 |
+
"loss": 0.6518,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.43821209465381245,
|
| 356 |
+
"grad_norm": 1.755315899848938,
|
| 357 |
+
"learning_rate": 8.979283820908174e-05,
|
| 358 |
+
"loss": 0.6145,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.4469763365468887,
|
| 363 |
+
"grad_norm": 1.6948930025100708,
|
| 364 |
+
"learning_rate": 8.939666566882821e-05,
|
| 365 |
+
"loss": 0.6809,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.45574057843996496,
|
| 370 |
+
"grad_norm": 1.9926345348358154,
|
| 371 |
+
"learning_rate": 8.899397417405442e-05,
|
| 372 |
+
"loss": 0.6811,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.4645048203330412,
|
| 377 |
+
"grad_norm": 2.0141072273254395,
|
| 378 |
+
"learning_rate": 8.858484004400496e-05,
|
| 379 |
+
"loss": 0.6326,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.4732690622261174,
|
| 384 |
+
"grad_norm": 1.5361511707305908,
|
| 385 |
+
"learning_rate": 8.816934081895105e-05,
|
| 386 |
+
"loss": 0.6316,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.4820333041191937,
|
| 391 |
+
"grad_norm": 2.0940563678741455,
|
| 392 |
+
"learning_rate": 8.774755524549503e-05,
|
| 393 |
+
"loss": 0.7468,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.49079754601226994,
|
| 398 |
+
"grad_norm": 2.6678926944732666,
|
| 399 |
+
"learning_rate": 8.731956326164591e-05,
|
| 400 |
+
"loss": 0.6853,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.49956178790534617,
|
| 405 |
+
"grad_norm": 2.2798871994018555,
|
| 406 |
+
"learning_rate": 8.688544598166935e-05,
|
| 407 |
+
"loss": 0.6488,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5083260297984225,
|
| 412 |
+
"grad_norm": 2.1741902828216553,
|
| 413 |
+
"learning_rate": 8.644528568071472e-05,
|
| 414 |
+
"loss": 0.7044,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.5170902716914987,
|
| 419 |
+
"grad_norm": 1.9935109615325928,
|
| 420 |
+
"learning_rate": 8.599916577922198e-05,
|
| 421 |
+
"loss": 0.6689,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.5258545135845749,
|
| 426 |
+
"grad_norm": 2.2509982585906982,
|
| 427 |
+
"learning_rate": 8.554717082711164e-05,
|
| 428 |
+
"loss": 0.6697,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5346187554776511,
|
| 433 |
+
"grad_norm": 1.5349043607711792,
|
| 434 |
+
"learning_rate": 8.508938648776062e-05,
|
| 435 |
+
"loss": 0.6741,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5433829973707275,
|
| 440 |
+
"grad_norm": 1.8080490827560425,
|
| 441 |
+
"learning_rate": 8.462589952176709e-05,
|
| 442 |
+
"loss": 0.6448,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5521472392638037,
|
| 447 |
+
"grad_norm": 1.4415456056594849,
|
| 448 |
+
"learning_rate": 8.415679777050735e-05,
|
| 449 |
+
"loss": 0.6585,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.5609114811568799,
|
| 454 |
+
"grad_norm": 1.9184411764144897,
|
| 455 |
+
"learning_rate": 8.368217013948786e-05,
|
| 456 |
+
"loss": 0.6934,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.5696757230499562,
|
| 461 |
+
"grad_norm": 2.0871620178222656,
|
| 462 |
+
"learning_rate": 8.320210658149562e-05,
|
| 463 |
+
"loss": 0.642,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.5784399649430324,
|
| 468 |
+
"grad_norm": 1.835964322090149,
|
| 469 |
+
"learning_rate": 8.271669807955007e-05,
|
| 470 |
+
"loss": 0.7067,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.5872042068361086,
|
| 475 |
+
"grad_norm": 2.068668842315674,
|
| 476 |
+
"learning_rate": 8.222603662965974e-05,
|
| 477 |
+
"loss": 0.6759,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.595968448729185,
|
| 482 |
+
"grad_norm": 1.837086796760559,
|
| 483 |
+
"learning_rate": 8.173021522338687e-05,
|
| 484 |
+
"loss": 0.6301,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.6047326906222612,
|
| 489 |
+
"grad_norm": 1.995373010635376,
|
| 490 |
+
"learning_rate": 8.122932783022342e-05,
|
| 491 |
+
"loss": 0.688,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.6134969325153374,
|
| 496 |
+
"grad_norm": 1.736932635307312,
|
| 497 |
+
"learning_rate": 8.072346937978168e-05,
|
| 498 |
+
"loss": 0.6538,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.6222611744084137,
|
| 503 |
+
"grad_norm": 2.074141025543213,
|
| 504 |
+
"learning_rate": 8.02127357438029e-05,
|
| 505 |
+
"loss": 0.7275,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.6310254163014899,
|
| 510 |
+
"grad_norm": 2.758039951324463,
|
| 511 |
+
"learning_rate": 7.969722371798753e-05,
|
| 512 |
+
"loss": 0.7139,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6397896581945661,
|
| 517 |
+
"grad_norm": 1.8415740728378296,
|
| 518 |
+
"learning_rate": 7.917703100365005e-05,
|
| 519 |
+
"loss": 0.5433,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.6485539000876425,
|
| 524 |
+
"grad_norm": 2.0133821964263916,
|
| 525 |
+
"learning_rate": 7.865225618920248e-05,
|
| 526 |
+
"loss": 0.6369,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.6573181419807187,
|
| 531 |
+
"grad_norm": 1.943287968635559,
|
| 532 |
+
"learning_rate": 7.812299873146955e-05,
|
| 533 |
+
"loss": 0.6168,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.6660823838737949,
|
| 538 |
+
"grad_norm": 1.7971194982528687,
|
| 539 |
+
"learning_rate": 7.758935893683939e-05,
|
| 540 |
+
"loss": 0.6518,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.6748466257668712,
|
| 545 |
+
"grad_norm": 1.5026414394378662,
|
| 546 |
+
"learning_rate": 7.705143794225315e-05,
|
| 547 |
+
"loss": 0.6498,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.6836108676599474,
|
| 552 |
+
"grad_norm": 1.8013416528701782,
|
| 553 |
+
"learning_rate": 7.65093376960372e-05,
|
| 554 |
+
"loss": 0.6598,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.6923751095530236,
|
| 559 |
+
"grad_norm": 1.8364579677581787,
|
| 560 |
+
"learning_rate": 7.596316093858172e-05,
|
| 561 |
+
"loss": 0.6934,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.7011393514461,
|
| 566 |
+
"grad_norm": 1.7535738945007324,
|
| 567 |
+
"learning_rate": 7.541301118286894e-05,
|
| 568 |
+
"loss": 0.5668,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.7099035933391762,
|
| 573 |
+
"grad_norm": 1.9401272535324097,
|
| 574 |
+
"learning_rate": 7.485899269485506e-05,
|
| 575 |
+
"loss": 0.6852,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.7186678352322524,
|
| 580 |
+
"grad_norm": 1.886604905128479,
|
| 581 |
+
"learning_rate": 7.430121047370955e-05,
|
| 582 |
+
"loss": 0.6845,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.7274320771253286,
|
| 587 |
+
"grad_norm": 1.9620131254196167,
|
| 588 |
+
"learning_rate": 7.37397702319153e-05,
|
| 589 |
+
"loss": 0.6344,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.7361963190184049,
|
| 594 |
+
"grad_norm": 1.949866771697998,
|
| 595 |
+
"learning_rate": 7.32314343370074e-05,
|
| 596 |
+
"loss": 0.6247,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.7449605609114811,
|
| 601 |
+
"grad_norm": 1.5609701871871948,
|
| 602 |
+
"learning_rate": 7.266333756059938e-05,
|
| 603 |
+
"loss": 0.604,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.7537248028045574,
|
| 608 |
+
"grad_norm": 2.5070557594299316,
|
| 609 |
+
"learning_rate": 7.209189317790467e-05,
|
| 610 |
+
"loss": 0.6262,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.7624890446976337,
|
| 615 |
+
"grad_norm": 1.672145128250122,
|
| 616 |
+
"learning_rate": 7.151720949069814e-05,
|
| 617 |
+
"loss": 0.6188,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.7712532865907099,
|
| 622 |
+
"grad_norm": 1.456264615058899,
|
| 623 |
+
"learning_rate": 7.093939541467697e-05,
|
| 624 |
+
"loss": 0.6026,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.7800175284837861,
|
| 629 |
+
"grad_norm": 1.967781901359558,
|
| 630 |
+
"learning_rate": 7.035856045881851e-05,
|
| 631 |
+
"loss": 0.6649,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.7887817703768624,
|
| 636 |
+
"grad_norm": 1.6206531524658203,
|
| 637 |
+
"learning_rate": 6.977481470462593e-05,
|
| 638 |
+
"loss": 0.6018,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.7975460122699386,
|
| 643 |
+
"grad_norm": 2.240525960922241,
|
| 644 |
+
"learning_rate": 6.918826878526527e-05,
|
| 645 |
+
"loss": 0.5912,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.8063102541630149,
|
| 650 |
+
"grad_norm": 1.8358842134475708,
|
| 651 |
+
"learning_rate": 6.859903386459781e-05,
|
| 652 |
+
"loss": 0.6435,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.8150744960560912,
|
| 657 |
+
"grad_norm": 2.574652671813965,
|
| 658 |
+
"learning_rate": 6.80072216161121e-05,
|
| 659 |
+
"loss": 0.6789,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.8238387379491674,
|
| 664 |
+
"grad_norm": 1.668445348739624,
|
| 665 |
+
"learning_rate": 6.741294420175927e-05,
|
| 666 |
+
"loss": 0.6103,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.8326029798422436,
|
| 671 |
+
"grad_norm": 1.7513461112976074,
|
| 672 |
+
"learning_rate": 6.681631425069566e-05,
|
| 673 |
+
"loss": 0.5696,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.8413672217353199,
|
| 678 |
+
"grad_norm": 2.508164405822754,
|
| 679 |
+
"learning_rate": 6.621744483793715e-05,
|
| 680 |
+
"loss": 0.653,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.8501314636283961,
|
| 685 |
+
"grad_norm": 1.9786324501037598,
|
| 686 |
+
"learning_rate": 6.56164494629288e-05,
|
| 687 |
+
"loss": 0.635,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.8588957055214724,
|
| 692 |
+
"grad_norm": 1.8781009912490845,
|
| 693 |
+
"learning_rate": 6.501344202803414e-05,
|
| 694 |
+
"loss": 0.7071,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.8676599474145487,
|
| 699 |
+
"grad_norm": 1.819648027420044,
|
| 700 |
+
"learning_rate": 6.440853681694801e-05,
|
| 701 |
+
"loss": 0.5952,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.8764241893076249,
|
| 706 |
+
"grad_norm": 1.8697113990783691,
|
| 707 |
+
"learning_rate": 6.380184847303727e-05,
|
| 708 |
+
"loss": 0.6179,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.8851884312007011,
|
| 713 |
+
"grad_norm": 1.6153459548950195,
|
| 714 |
+
"learning_rate": 6.319349197761317e-05,
|
| 715 |
+
"loss": 0.6071,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.8939526730937774,
|
| 720 |
+
"grad_norm": 1.5980494022369385,
|
| 721 |
+
"learning_rate": 6.25835826281398e-05,
|
| 722 |
+
"loss": 0.588,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.9027169149868537,
|
| 727 |
+
"grad_norm": 1.509323239326477,
|
| 728 |
+
"learning_rate": 6.197223601638266e-05,
|
| 729 |
+
"loss": 0.6636,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.9114811568799299,
|
| 734 |
+
"grad_norm": 1.91567862033844,
|
| 735 |
+
"learning_rate": 6.135956800650128e-05,
|
| 736 |
+
"loss": 0.6727,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.9202453987730062,
|
| 741 |
+
"grad_norm": 1.7963509559631348,
|
| 742 |
+
"learning_rate": 6.074569471309032e-05,
|
| 743 |
+
"loss": 0.5768,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.9290096406660824,
|
| 748 |
+
"grad_norm": 1.8217496871948242,
|
| 749 |
+
"learning_rate": 6.013073247917326e-05,
|
| 750 |
+
"loss": 0.6243,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.9377738825591586,
|
| 755 |
+
"grad_norm": 1.648887276649475,
|
| 756 |
+
"learning_rate": 5.951479785415266e-05,
|
| 757 |
+
"loss": 0.6717,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.9465381244522348,
|
| 762 |
+
"grad_norm": 1.668747067451477,
|
| 763 |
+
"learning_rate": 5.889800757172146e-05,
|
| 764 |
+
"loss": 0.5823,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.9553023663453112,
|
| 769 |
+
"grad_norm": 2.224945545196533,
|
| 770 |
+
"learning_rate": 5.8280478527739235e-05,
|
| 771 |
+
"loss": 0.6203,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.9640666082383874,
|
| 776 |
+
"grad_norm": NaN,
|
| 777 |
+
"learning_rate": 5.7724167474641534e-05,
|
| 778 |
+
"loss": 0.6271,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.9728308501314636,
|
| 783 |
+
"grad_norm": 2.2267634868621826,
|
| 784 |
+
"learning_rate": 5.7105557315385284e-05,
|
| 785 |
+
"loss": 0.6614,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.9815950920245399,
|
| 790 |
+
"grad_norm": 1.7963807582855225,
|
| 791 |
+
"learning_rate": 5.6486548104880555e-05,
|
| 792 |
+
"loss": 0.7113,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.9903593339176161,
|
| 797 |
+
"grad_norm": 2.0616729259490967,
|
| 798 |
+
"learning_rate": 5.586725715952452e-05,
|
| 799 |
+
"loss": 0.6077,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.9991235758106923,
|
| 804 |
+
"grad_norm": 1.7366951704025269,
|
| 805 |
+
"learning_rate": 5.5247801849109526e-05,
|
| 806 |
+
"loss": 0.634,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.0078878177037687,
|
| 811 |
+
"grad_norm": 1.4753001928329468,
|
| 812 |
+
"learning_rate": 5.462829957457888e-05,
|
| 813 |
+
"loss": 0.44,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.016652059596845,
|
| 818 |
+
"grad_norm": 1.493257999420166,
|
| 819 |
+
"learning_rate": 5.400886774577667e-05,
|
| 820 |
+
"loss": 0.3635,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.0254163014899211,
|
| 825 |
+
"grad_norm": 1.7157448530197144,
|
| 826 |
+
"learning_rate": 5.338962375919589e-05,
|
| 827 |
+
"loss": 0.3525,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.0341805433829974,
|
| 832 |
+
"grad_norm": 1.7190569639205933,
|
| 833 |
+
"learning_rate": 5.277068497572914e-05,
|
| 834 |
+
"loss": 0.3751,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.0429447852760736,
|
| 839 |
+
"grad_norm": 1.5113599300384521,
|
| 840 |
+
"learning_rate": 5.215216869842604e-05,
|
| 841 |
+
"loss": 0.3857,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.0517090271691498,
|
| 846 |
+
"grad_norm": 1.6809605360031128,
|
| 847 |
+
"learning_rate": 5.1534192150261676e-05,
|
| 848 |
+
"loss": 0.3611,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.060473269062226,
|
| 853 |
+
"grad_norm": 1.6623985767364502,
|
| 854 |
+
"learning_rate": 5.091687245192006e-05,
|
| 855 |
+
"loss": 0.3558,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.0692375109553023,
|
| 860 |
+
"grad_norm": 2.009783983230591,
|
| 861 |
+
"learning_rate": 5.030032659959722e-05,
|
| 862 |
+
"loss": 0.3654,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.0780017528483785,
|
| 867 |
+
"grad_norm": 1.6117240190505981,
|
| 868 |
+
"learning_rate": 4.968467144282759e-05,
|
| 869 |
+
"loss": 0.3715,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.086765994741455,
|
| 874 |
+
"grad_norm": 1.3283199071884155,
|
| 875 |
+
"learning_rate": 4.9070023662338523e-05,
|
| 876 |
+
"loss": 0.3736,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.0955302366345312,
|
| 881 |
+
"grad_norm": 1.8791778087615967,
|
| 882 |
+
"learning_rate": 4.8456499747936465e-05,
|
| 883 |
+
"loss": 0.4082,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.1042944785276074,
|
| 888 |
+
"grad_norm": 1.7714508771896362,
|
| 889 |
+
"learning_rate": 4.7844215976429576e-05,
|
| 890 |
+
"loss": 0.3743,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.1130587204206837,
|
| 895 |
+
"grad_norm": 1.773903489112854,
|
| 896 |
+
"learning_rate": 4.723328838959057e-05,
|
| 897 |
+
"loss": 0.3858,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.1218229623137599,
|
| 902 |
+
"grad_norm": 1.4871599674224854,
|
| 903 |
+
"learning_rate": 4.662383277216418e-05,
|
| 904 |
+
"loss": 0.3486,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.1305872042068361,
|
| 909 |
+
"grad_norm": 1.3975831270217896,
|
| 910 |
+
"learning_rate": 4.601596462992326e-05,
|
| 911 |
+
"loss": 0.3657,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.1393514460999123,
|
| 916 |
+
"grad_norm": 1.5370299816131592,
|
| 917 |
+
"learning_rate": 4.540979916777783e-05,
|
| 918 |
+
"loss": 0.3371,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.1481156879929886,
|
| 923 |
+
"grad_norm": 1.5495011806488037,
|
| 924 |
+
"learning_rate": 4.480545126794115e-05,
|
| 925 |
+
"loss": 0.3357,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.1568799298860648,
|
| 930 |
+
"grad_norm": 1.694589376449585,
|
| 931 |
+
"learning_rate": 4.420303546815678e-05,
|
| 932 |
+
"loss": 0.3645,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.165644171779141,
|
| 937 |
+
"grad_norm": 1.985825777053833,
|
| 938 |
+
"learning_rate": 4.360266593999124e-05,
|
| 939 |
+
"loss": 0.3546,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.1744084136722173,
|
| 944 |
+
"grad_norm": 1.5808460712432861,
|
| 945 |
+
"learning_rate": 4.300445646719573e-05,
|
| 946 |
+
"loss": 0.3638,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.1831726555652935,
|
| 951 |
+
"grad_norm": 1.7759652137756348,
|
| 952 |
+
"learning_rate": 4.240852042414162e-05,
|
| 953 |
+
"loss": 0.4059,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.19193689745837,
|
| 958 |
+
"grad_norm": 1.7563869953155518,
|
| 959 |
+
"learning_rate": 4.181497075433334e-05,
|
| 960 |
+
"loss": 0.3446,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.2007011393514462,
|
| 965 |
+
"grad_norm": 1.9143909215927124,
|
| 966 |
+
"learning_rate": 4.1223919949003045e-05,
|
| 967 |
+
"loss": 0.3487,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.2094653812445224,
|
| 972 |
+
"grad_norm": 1.8407511711120605,
|
| 973 |
+
"learning_rate": 4.0635480025790926e-05,
|
| 974 |
+
"loss": 0.3649,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.2182296231375986,
|
| 979 |
+
"grad_norm": 1.7706880569458008,
|
| 980 |
+
"learning_rate": 4.0049762507515355e-05,
|
| 981 |
+
"loss": 0.3612,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.2269938650306749,
|
| 986 |
+
"grad_norm": 1.0870561599731445,
|
| 987 |
+
"learning_rate": 3.9466878401036686e-05,
|
| 988 |
+
"loss": 0.3401,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.235758106923751,
|
| 993 |
+
"grad_norm": 1.91828453540802,
|
| 994 |
+
"learning_rate": 3.8886938176219024e-05,
|
| 995 |
+
"loss": 0.3327,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.2445223488168273,
|
| 1000 |
+
"grad_norm": 1.61056649684906,
|
| 1001 |
+
"learning_rate": 3.8310051744993514e-05,
|
| 1002 |
+
"loss": 0.3386,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.2532865907099036,
|
| 1007 |
+
"grad_norm": 2.071869373321533,
|
| 1008 |
+
"learning_rate": 3.773632844052767e-05,
|
| 1009 |
+
"loss": 0.363,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.2620508326029798,
|
| 1014 |
+
"grad_norm": 1.671288251876831,
|
| 1015 |
+
"learning_rate": 3.7165876996504125e-05,
|
| 1016 |
+
"loss": 0.3828,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.270815074496056,
|
| 1021 |
+
"grad_norm": 1.8811005353927612,
|
| 1022 |
+
"learning_rate": 3.659880552651317e-05,
|
| 1023 |
+
"loss": 0.3551,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.2795793163891322,
|
| 1028 |
+
"grad_norm": 1.5208740234375,
|
| 1029 |
+
"learning_rate": 3.6035221503562775e-05,
|
| 1030 |
+
"loss": 0.3566,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.2883435582822087,
|
| 1035 |
+
"grad_norm": 1.7736235857009888,
|
| 1036 |
+
"learning_rate": 3.547523173970989e-05,
|
| 1037 |
+
"loss": 0.3629,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.2971078001752847,
|
| 1042 |
+
"grad_norm": 1.6049748659133911,
|
| 1043 |
+
"learning_rate": 3.491894236581728e-05,
|
| 1044 |
+
"loss": 0.3359,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.3058720420683612,
|
| 1049 |
+
"grad_norm": 1.4260120391845703,
|
| 1050 |
+
"learning_rate": 3.436645881143918e-05,
|
| 1051 |
+
"loss": 0.3758,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.3146362839614374,
|
| 1056 |
+
"grad_norm": 1.9971890449523926,
|
| 1057 |
+
"learning_rate": 3.3817885784839986e-05,
|
| 1058 |
+
"loss": 0.314,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.3234005258545136,
|
| 1063 |
+
"grad_norm": 1.7832164764404297,
|
| 1064 |
+
"learning_rate": 3.327332725314974e-05,
|
| 1065 |
+
"loss": 0.3901,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.3321647677475899,
|
| 1070 |
+
"grad_norm": 1.6533173322677612,
|
| 1071 |
+
"learning_rate": 3.273288642265985e-05,
|
| 1072 |
+
"loss": 0.3324,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.340929009640666,
|
| 1077 |
+
"grad_norm": 1.5945855379104614,
|
| 1078 |
+
"learning_rate": 3.2196665719263266e-05,
|
| 1079 |
+
"loss": 0.3435,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.3496932515337423,
|
| 1084 |
+
"grad_norm": 1.51680588722229,
|
| 1085 |
+
"learning_rate": 3.166476676904235e-05,
|
| 1086 |
+
"loss": 0.3714,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.3584574934268185,
|
| 1091 |
+
"grad_norm": 1.6684399843215942,
|
| 1092 |
+
"learning_rate": 3.113729037900843e-05,
|
| 1093 |
+
"loss": 0.333,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.3672217353198948,
|
| 1098 |
+
"grad_norm": 1.5105247497558594,
|
| 1099 |
+
"learning_rate": 3.0614336517996576e-05,
|
| 1100 |
+
"loss": 0.3615,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.375985977212971,
|
| 1105 |
+
"grad_norm": 2.113157272338867,
|
| 1106 |
+
"learning_rate": 3.0096004297719205e-05,
|
| 1107 |
+
"loss": 0.3002,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.3847502191060475,
|
| 1112 |
+
"grad_norm": 1.1173641681671143,
|
| 1113 |
+
"learning_rate": 2.958239195398217e-05,
|
| 1114 |
+
"loss": 0.3571,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 1.3935144609991235,
|
| 1119 |
+
"grad_norm": 1.4548070430755615,
|
| 1120 |
+
"learning_rate": 2.90735968280668e-05,
|
| 1121 |
+
"loss": 0.3252,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 1.4022787028922,
|
| 1126 |
+
"grad_norm": 1.7521406412124634,
|
| 1127 |
+
"learning_rate": 2.8569715348281547e-05,
|
| 1128 |
+
"loss": 0.3409,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.4110429447852761,
|
| 1133 |
+
"grad_norm": 1.7292704582214355,
|
| 1134 |
+
"learning_rate": 2.807084301168652e-05,
|
| 1135 |
+
"loss": 0.3282,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 1.4198071866783524,
|
| 1140 |
+
"grad_norm": 1.7453776597976685,
|
| 1141 |
+
"learning_rate": 2.7577074365994747e-05,
|
| 1142 |
+
"loss": 0.3217,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 1.4285714285714286,
|
| 1147 |
+
"grad_norm": 2.02482271194458,
|
| 1148 |
+
"learning_rate": 2.70885029916531e-05,
|
| 1149 |
+
"loss": 0.3355,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.4373356704645048,
|
| 1154 |
+
"grad_norm": 1.927150845527649,
|
| 1155 |
+
"learning_rate": 2.660522148410675e-05,
|
| 1156 |
+
"loss": 0.3528,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 1.446099912357581,
|
| 1161 |
+
"grad_norm": 1.5125665664672852,
|
| 1162 |
+
"learning_rate": 2.6127321436250117e-05,
|
| 1163 |
+
"loss": 0.3355,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 1.4548641542506573,
|
| 1168 |
+
"grad_norm": 1.771646499633789,
|
| 1169 |
+
"learning_rate": 2.565489342106805e-05,
|
| 1170 |
+
"loss": 0.3375,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 1.4636283961437335,
|
| 1175 |
+
"grad_norm": 2.540931463241577,
|
| 1176 |
+
"learning_rate": 2.518802697447003e-05,
|
| 1177 |
+
"loss": 0.3601,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 1.4723926380368098,
|
| 1182 |
+
"grad_norm": 1.5853700637817383,
|
| 1183 |
+
"learning_rate": 2.472681057832121e-05,
|
| 1184 |
+
"loss": 0.3207,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 1.481156879929886,
|
| 1189 |
+
"grad_norm": 2.11691951751709,
|
| 1190 |
+
"learning_rate": 2.427133164367296e-05,
|
| 1191 |
+
"loss": 0.3357,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 1.4899211218229622,
|
| 1196 |
+
"grad_norm": 1.55061674118042,
|
| 1197 |
+
"learning_rate": 2.3821676494196572e-05,
|
| 1198 |
+
"loss": 0.3314,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 1.4986853637160387,
|
| 1203 |
+
"grad_norm": 1.5113292932510376,
|
| 1204 |
+
"learning_rate": 2.3377930349822856e-05,
|
| 1205 |
+
"loss": 0.3162,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.5074496056091147,
|
| 1210 |
+
"grad_norm": 2.0338950157165527,
|
| 1211 |
+
"learning_rate": 2.2940177310591113e-05,
|
| 1212 |
+
"loss": 0.3693,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 1.5162138475021911,
|
| 1217 |
+
"grad_norm": 2.0127060413360596,
|
| 1218 |
+
"learning_rate": 2.250850034071016e-05,
|
| 1219 |
+
"loss": 0.324,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 1.5249780893952674,
|
| 1224 |
+
"grad_norm": 1.9195280075073242,
|
| 1225 |
+
"learning_rate": 2.20829812528348e-05,
|
| 1226 |
+
"loss": 0.3559,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 1.5337423312883436,
|
| 1231 |
+
"grad_norm": 1.6400375366210938,
|
| 1232 |
+
"learning_rate": 2.1663700692560373e-05,
|
| 1233 |
+
"loss": 0.3371,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 1.5425065731814198,
|
| 1238 |
+
"grad_norm": 1.9201463460922241,
|
| 1239 |
+
"learning_rate": 2.1250738123138665e-05,
|
| 1240 |
+
"loss": 0.3536,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 1.551270815074496,
|
| 1245 |
+
"grad_norm": 1.8780487775802612,
|
| 1246 |
+
"learning_rate": 2.084417181041769e-05,
|
| 1247 |
+
"loss": 0.3829,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.5600350569675723,
|
| 1252 |
+
"grad_norm": 1.8463397026062012,
|
| 1253 |
+
"learning_rate": 2.0444078808008655e-05,
|
| 1254 |
+
"loss": 0.3431,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 1.5687992988606485,
|
| 1259 |
+
"grad_norm": 1.8080838918685913,
|
| 1260 |
+
"learning_rate": 2.005053494268241e-05,
|
| 1261 |
+
"loss": 0.3748,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.577563540753725,
|
| 1266 |
+
"grad_norm": 1.9387633800506592,
|
| 1267 |
+
"learning_rate": 1.9663614799998635e-05,
|
| 1268 |
+
"loss": 0.3586,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 1.586327782646801,
|
| 1273 |
+
"grad_norm": 1.7084999084472656,
|
| 1274 |
+
"learning_rate": 1.928339171017015e-05,
|
| 1275 |
+
"loss": 0.357,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 1.5950920245398774,
|
| 1280 |
+
"grad_norm": 2.1515519618988037,
|
| 1281 |
+
"learning_rate": 1.8909937734165107e-05,
|
| 1282 |
+
"loss": 0.3333,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.6038562664329534,
|
| 1287 |
+
"grad_norm": 1.4635021686553955,
|
| 1288 |
+
"learning_rate": 1.8543323650049864e-05,
|
| 1289 |
+
"loss": 0.3337,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 1.6126205083260299,
|
| 1294 |
+
"grad_norm": 1.7444740533828735,
|
| 1295 |
+
"learning_rate": 1.8183618939574904e-05,
|
| 1296 |
+
"loss": 0.3686,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 1.6213847502191059,
|
| 1301 |
+
"grad_norm": 1.587889552116394,
|
| 1302 |
+
"learning_rate": 1.7830891775006396e-05,
|
| 1303 |
+
"loss": 0.3167,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 1.6301489921121823,
|
| 1308 |
+
"grad_norm": 1.9329102039337158,
|
| 1309 |
+
"learning_rate": 1.748520900620609e-05,
|
| 1310 |
+
"loss": 0.3349,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 1.6389132340052586,
|
| 1315 |
+
"grad_norm": 1.610293984413147,
|
| 1316 |
+
"learning_rate": 1.714663614796167e-05,
|
| 1317 |
+
"loss": 0.3841,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 1.6476774758983348,
|
| 1322 |
+
"grad_norm": 1.9500341415405273,
|
| 1323 |
+
"learning_rate": 1.6815237367570197e-05,
|
| 1324 |
+
"loss": 0.3575,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.656441717791411,
|
| 1329 |
+
"grad_norm": 1.717809796333313,
|
| 1330 |
+
"learning_rate": 1.6491075472677016e-05,
|
| 1331 |
+
"loss": 0.2917,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 1.6652059596844873,
|
| 1336 |
+
"grad_norm": 1.2370789051055908,
|
| 1337 |
+
"learning_rate": 1.6174211899372175e-05,
|
| 1338 |
+
"loss": 0.3535,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.6739702015775635,
|
| 1343 |
+
"grad_norm": 1.6636922359466553,
|
| 1344 |
+
"learning_rate": 1.5864706700546955e-05,
|
| 1345 |
+
"loss": 0.3242,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.6827344434706397,
|
| 1350 |
+
"grad_norm": 1.3746954202651978,
|
| 1351 |
+
"learning_rate": 1.5562618534512428e-05,
|
| 1352 |
+
"loss": 0.3463,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 1.6914986853637162,
|
| 1357 |
+
"grad_norm": 1.9031116962432861,
|
| 1358 |
+
"learning_rate": 1.5268004653882406e-05,
|
| 1359 |
+
"loss": 0.3448,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 1.7002629272567922,
|
| 1364 |
+
"grad_norm": 1.8231595754623413,
|
| 1365 |
+
"learning_rate": 1.4980920894722692e-05,
|
| 1366 |
+
"loss": 0.3327,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.7090271691498686,
|
| 1371 |
+
"grad_norm": 2.2330541610717773,
|
| 1372 |
+
"learning_rate": 1.4701421665969001e-05,
|
| 1373 |
+
"loss": 0.344,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 1.7177914110429446,
|
| 1378 |
+
"grad_norm": 1.5624206066131592,
|
| 1379 |
+
"learning_rate": 1.442955993911505e-05,
|
| 1380 |
+
"loss": 0.3194,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 1.726555652936021,
|
| 1385 |
+
"grad_norm": 1.7255851030349731,
|
| 1386 |
+
"learning_rate": 1.4165387238173399e-05,
|
| 1387 |
+
"loss": 0.3122,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 1.7353198948290973,
|
| 1392 |
+
"grad_norm": 1.7800745964050293,
|
| 1393 |
+
"learning_rate": 1.3908953629910376e-05,
|
| 1394 |
+
"loss": 0.3005,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 1.7440841367221736,
|
| 1399 |
+
"grad_norm": 2.472658395767212,
|
| 1400 |
+
"learning_rate": 1.3660307714357338e-05,
|
| 1401 |
+
"loss": 0.3571,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 1.7528483786152498,
|
| 1406 |
+
"grad_norm": 1.6620064973831177,
|
| 1407 |
+
"learning_rate": 1.3419496615599805e-05,
|
| 1408 |
+
"loss": 0.3522,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 1.761612620508326,
|
| 1413 |
+
"grad_norm": 1.5490673780441284,
|
| 1414 |
+
"learning_rate": 1.318656597284643e-05,
|
| 1415 |
+
"loss": 0.2967,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 1.7703768624014022,
|
| 1420 |
+
"grad_norm": 2.1536474227905273,
|
| 1421 |
+
"learning_rate": 1.2961559931779257e-05,
|
| 1422 |
+
"loss": 0.3703,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 1.7791411042944785,
|
| 1427 |
+
"grad_norm": 1.490907073020935,
|
| 1428 |
+
"learning_rate": 1.274452113618716e-05,
|
| 1429 |
+
"loss": 0.3076,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 1.787905346187555,
|
| 1434 |
+
"grad_norm": 1.4357167482376099,
|
| 1435 |
+
"learning_rate": 1.2535490719883835e-05,
|
| 1436 |
+
"loss": 0.3265,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 1.796669588080631,
|
| 1441 |
+
"grad_norm": 1.5666704177856445,
|
| 1442 |
+
"learning_rate": 1.233450829891203e-05,
|
| 1443 |
+
"loss": 0.324,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 1.8054338299737074,
|
| 1448 |
+
"grad_norm": 1.3453813791275024,
|
| 1449 |
+
"learning_rate": 1.2141611964035366e-05,
|
| 1450 |
+
"loss": 0.2977,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 1.8141980718667834,
|
| 1455 |
+
"grad_norm": 1.782593011856079,
|
| 1456 |
+
"learning_rate": 1.195683827351931e-05,
|
| 1457 |
+
"loss": 0.2795,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 1.8229623137598598,
|
| 1462 |
+
"grad_norm": 1.9441533088684082,
|
| 1463 |
+
"learning_rate": 1.1780222246202494e-05,
|
| 1464 |
+
"loss": 0.3166,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 1.831726555652936,
|
| 1469 |
+
"grad_norm": 1.778911828994751,
|
| 1470 |
+
"learning_rate": 1.1611797354859892e-05,
|
| 1471 |
+
"loss": 0.3493,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 1.8404907975460123,
|
| 1476 |
+
"grad_norm": 1.7141963243484497,
|
| 1477 |
+
"learning_rate": 1.145159551985894e-05,
|
| 1478 |
+
"loss": 0.3313,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 1.8492550394390885,
|
| 1483 |
+
"grad_norm": 3.3165926933288574,
|
| 1484 |
+
"learning_rate": 1.1299647103109908e-05,
|
| 1485 |
+
"loss": 0.356,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 1.8580192813321648,
|
| 1490 |
+
"grad_norm": 1.787851095199585,
|
| 1491 |
+
"learning_rate": 1.11559809023116e-05,
|
| 1492 |
+
"loss": 0.3219,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 1.866783523225241,
|
| 1497 |
+
"grad_norm": 1.7090057134628296,
|
| 1498 |
+
"learning_rate": 1.1020624145493572e-05,
|
| 1499 |
+
"loss": 0.3445,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 1.8755477651183172,
|
| 1504 |
+
"grad_norm": 1.5748744010925293,
|
| 1505 |
+
"learning_rate": 1.0893602485855766e-05,
|
| 1506 |
+
"loss": 0.3253,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 1.8843120070113937,
|
| 1511 |
+
"grad_norm": 1.9979685544967651,
|
| 1512 |
+
"learning_rate": 1.0774939996906644e-05,
|
| 1513 |
+
"loss": 0.3177,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 1.8930762489044697,
|
| 1518 |
+
"grad_norm": 1.936296820640564,
|
| 1519 |
+
"learning_rate": 1.0664659167900723e-05,
|
| 1520 |
+
"loss": 0.3304,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 1.9018404907975461,
|
| 1525 |
+
"grad_norm": 1.8121618032455444,
|
| 1526 |
+
"learning_rate": 1.0562780899576344e-05,
|
| 1527 |
+
"loss": 0.3493,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 1.9106047326906221,
|
| 1532 |
+
"grad_norm": 1.6870180368423462,
|
| 1533 |
+
"learning_rate": 1.046932450019448e-05,
|
| 1534 |
+
"loss": 0.328,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 1.9193689745836986,
|
| 1539 |
+
"grad_norm": 2.0619869232177734,
|
| 1540 |
+
"learning_rate": 1.0384307681879428e-05,
|
| 1541 |
+
"loss": 0.3845,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 1.9281332164767746,
|
| 1546 |
+
"grad_norm": 2.1429038047790527,
|
| 1547 |
+
"learning_rate": 1.030774655726191e-05,
|
| 1548 |
+
"loss": 0.3143,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 1.936897458369851,
|
| 1553 |
+
"grad_norm": 1.9444646835327148,
|
| 1554 |
+
"learning_rate": 1.0239655636425374e-05,
|
| 1555 |
+
"loss": 0.3135,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 1.9456617002629273,
|
| 1560 |
+
"grad_norm": 1.6065791845321655,
|
| 1561 |
+
"learning_rate": 1.0180047824156011e-05,
|
| 1562 |
+
"loss": 0.3142,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 1.9544259421560035,
|
| 1567 |
+
"grad_norm": 2.215041160583496,
|
| 1568 |
+
"learning_rate": 1.0128934417497004e-05,
|
| 1569 |
+
"loss": 0.3234,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 1.9631901840490797,
|
| 1574 |
+
"grad_norm": 1.766499638557434,
|
| 1575 |
+
"learning_rate": 1.008632510360747e-05,
|
| 1576 |
+
"loss": 0.3395,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 1.971954425942156,
|
| 1581 |
+
"grad_norm": 2.355278491973877,
|
| 1582 |
+
"learning_rate": 1.0052227957926518e-05,
|
| 1583 |
+
"loss": 0.3476,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 1.9807186678352322,
|
| 1588 |
+
"grad_norm": 1.6923573017120361,
|
| 1589 |
+
"learning_rate": 1.0026649442642785e-05,
|
| 1590 |
+
"loss": 0.386,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 1.9894829097283084,
|
| 1595 |
+
"grad_norm": 1.454087495803833,
|
| 1596 |
+
"learning_rate": 1.0009594405469695e-05,
|
| 1597 |
+
"loss": 0.3059,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 1.9982471516213849,
|
| 1602 |
+
"grad_norm": 1.5868600606918335,
|
| 1603 |
+
"learning_rate": 1.0001066078726703e-05,
|
| 1604 |
+
"loss": 0.3474,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
}
|
| 1607 |
+
],
|
| 1608 |
+
"logging_steps": 10,
|
| 1609 |
+
"max_steps": 2282,
|
| 1610 |
+
"num_input_tokens_seen": 0,
|
| 1611 |
+
"num_train_epochs": 2,
|
| 1612 |
+
"save_steps": 500,
|
| 1613 |
+
"stateful_callbacks": {
|
| 1614 |
+
"TrainerControl": {
|
| 1615 |
+
"args": {
|
| 1616 |
+
"should_epoch_stop": false,
|
| 1617 |
+
"should_evaluate": false,
|
| 1618 |
+
"should_log": false,
|
| 1619 |
+
"should_save": true,
|
| 1620 |
+
"should_training_stop": true
|
| 1621 |
+
},
|
| 1622 |
+
"attributes": {}
|
| 1623 |
+
}
|
| 1624 |
+
},
|
| 1625 |
+
"total_flos": 2.7769146165323366e+17,
|
| 1626 |
+
"train_batch_size": 2,
|
| 1627 |
+
"trial_name": null,
|
| 1628 |
+
"trial_params": null
|
| 1629 |
+
}
|