Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +3004 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,3004 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 4230,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0070921985815602835,
|
| 14 |
+
"grad_norm": 10.550431251525879,
|
| 15 |
+
"learning_rate": 4.245283018867925e-06,
|
| 16 |
+
"loss": 0.9319,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.014184397163120567,
|
| 21 |
+
"grad_norm": 2.5195791721343994,
|
| 22 |
+
"learning_rate": 8.962264150943396e-06,
|
| 23 |
+
"loss": 0.3829,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02127659574468085,
|
| 28 |
+
"grad_norm": 1.7604764699935913,
|
| 29 |
+
"learning_rate": 1.3679245283018869e-05,
|
| 30 |
+
"loss": 0.245,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.028368794326241134,
|
| 35 |
+
"grad_norm": 1.5353305339813232,
|
| 36 |
+
"learning_rate": 1.839622641509434e-05,
|
| 37 |
+
"loss": 0.2284,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03546099290780142,
|
| 42 |
+
"grad_norm": 4.206876277923584,
|
| 43 |
+
"learning_rate": 2.3113207547169813e-05,
|
| 44 |
+
"loss": 0.1984,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0425531914893617,
|
| 49 |
+
"grad_norm": 2.1158673763275146,
|
| 50 |
+
"learning_rate": 2.7830188679245282e-05,
|
| 51 |
+
"loss": 0.1742,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.04964539007092199,
|
| 56 |
+
"grad_norm": 1.795837163925171,
|
| 57 |
+
"learning_rate": 3.254716981132075e-05,
|
| 58 |
+
"loss": 0.1932,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.05673758865248227,
|
| 63 |
+
"grad_norm": 1.7584210634231567,
|
| 64 |
+
"learning_rate": 3.7264150943396224e-05,
|
| 65 |
+
"loss": 0.1552,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.06382978723404255,
|
| 70 |
+
"grad_norm": 1.8413468599319458,
|
| 71 |
+
"learning_rate": 4.1981132075471704e-05,
|
| 72 |
+
"loss": 0.1198,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.07092198581560284,
|
| 77 |
+
"grad_norm": 0.7560085654258728,
|
| 78 |
+
"learning_rate": 4.669811320754717e-05,
|
| 79 |
+
"loss": 0.1096,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.07801418439716312,
|
| 84 |
+
"grad_norm": 0.9906244874000549,
|
| 85 |
+
"learning_rate": 5.141509433962265e-05,
|
| 86 |
+
"loss": 0.102,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0851063829787234,
|
| 91 |
+
"grad_norm": 0.693999707698822,
|
| 92 |
+
"learning_rate": 5.613207547169812e-05,
|
| 93 |
+
"loss": 0.0894,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.09219858156028368,
|
| 98 |
+
"grad_norm": 1.310461401939392,
|
| 99 |
+
"learning_rate": 6.084905660377359e-05,
|
| 100 |
+
"loss": 0.0774,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.09929078014184398,
|
| 105 |
+
"grad_norm": 1.3748232126235962,
|
| 106 |
+
"learning_rate": 6.556603773584906e-05,
|
| 107 |
+
"loss": 0.0769,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.10638297872340426,
|
| 112 |
+
"grad_norm": 1.2116252183914185,
|
| 113 |
+
"learning_rate": 7.028301886792453e-05,
|
| 114 |
+
"loss": 0.079,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.11347517730496454,
|
| 119 |
+
"grad_norm": 1.4632856845855713,
|
| 120 |
+
"learning_rate": 7.500000000000001e-05,
|
| 121 |
+
"loss": 0.0734,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.12056737588652482,
|
| 126 |
+
"grad_norm": 1.215768814086914,
|
| 127 |
+
"learning_rate": 7.971698113207548e-05,
|
| 128 |
+
"loss": 0.0623,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1276595744680851,
|
| 133 |
+
"grad_norm": 1.1149842739105225,
|
| 134 |
+
"learning_rate": 8.443396226415094e-05,
|
| 135 |
+
"loss": 0.0733,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1347517730496454,
|
| 140 |
+
"grad_norm": 0.8909507393836975,
|
| 141 |
+
"learning_rate": 8.915094339622641e-05,
|
| 142 |
+
"loss": 0.0748,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.14184397163120568,
|
| 147 |
+
"grad_norm": 0.9022103548049927,
|
| 148 |
+
"learning_rate": 9.386792452830188e-05,
|
| 149 |
+
"loss": 0.0739,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.14893617021276595,
|
| 154 |
+
"grad_norm": 0.7338738441467285,
|
| 155 |
+
"learning_rate": 9.858490566037736e-05,
|
| 156 |
+
"loss": 0.0568,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.15602836879432624,
|
| 161 |
+
"grad_norm": 0.751677393913269,
|
| 162 |
+
"learning_rate": 9.999925111542543e-05,
|
| 163 |
+
"loss": 0.0534,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.16312056737588654,
|
| 168 |
+
"grad_norm": 1.3503477573394775,
|
| 169 |
+
"learning_rate": 9.99955831633497e-05,
|
| 170 |
+
"loss": 0.0573,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.1702127659574468,
|
| 175 |
+
"grad_norm": 0.8441567420959473,
|
| 176 |
+
"learning_rate": 9.998885881749846e-05,
|
| 177 |
+
"loss": 0.0638,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.1773049645390071,
|
| 182 |
+
"grad_norm": 0.7064261436462402,
|
| 183 |
+
"learning_rate": 9.997907848895302e-05,
|
| 184 |
+
"loss": 0.0561,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.18439716312056736,
|
| 189 |
+
"grad_norm": 1.2360883951187134,
|
| 190 |
+
"learning_rate": 9.99662427756169e-05,
|
| 191 |
+
"loss": 0.0567,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.19148936170212766,
|
| 196 |
+
"grad_norm": 0.8929492235183716,
|
| 197 |
+
"learning_rate": 9.995035246217928e-05,
|
| 198 |
+
"loss": 0.0616,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.19858156028368795,
|
| 203 |
+
"grad_norm": 1.1510146856307983,
|
| 204 |
+
"learning_rate": 9.993140852006707e-05,
|
| 205 |
+
"loss": 0.0556,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.20567375886524822,
|
| 210 |
+
"grad_norm": 0.8707113265991211,
|
| 211 |
+
"learning_rate": 9.990941210738554e-05,
|
| 212 |
+
"loss": 0.0534,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2127659574468085,
|
| 217 |
+
"grad_norm": 0.6750063896179199,
|
| 218 |
+
"learning_rate": 9.988436456884744e-05,
|
| 219 |
+
"loss": 0.0532,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2198581560283688,
|
| 224 |
+
"grad_norm": 0.8051894903182983,
|
| 225 |
+
"learning_rate": 9.985626743569084e-05,
|
| 226 |
+
"loss": 0.0573,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.22695035460992907,
|
| 231 |
+
"grad_norm": 0.9641497135162354,
|
| 232 |
+
"learning_rate": 9.982512242558555e-05,
|
| 233 |
+
"loss": 0.0545,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.23404255319148937,
|
| 238 |
+
"grad_norm": 0.6739592552185059,
|
| 239 |
+
"learning_rate": 9.979093144252805e-05,
|
| 240 |
+
"loss": 0.05,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.24113475177304963,
|
| 245 |
+
"grad_norm": 0.5896672010421753,
|
| 246 |
+
"learning_rate": 9.975369657672514e-05,
|
| 247 |
+
"loss": 0.0459,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.24822695035460993,
|
| 252 |
+
"grad_norm": 0.721316397190094,
|
| 253 |
+
"learning_rate": 9.971342010446615e-05,
|
| 254 |
+
"loss": 0.0541,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.2553191489361702,
|
| 259 |
+
"grad_norm": 0.7230843305587769,
|
| 260 |
+
"learning_rate": 9.967010448798375e-05,
|
| 261 |
+
"loss": 0.0475,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.2624113475177305,
|
| 266 |
+
"grad_norm": 0.5599878430366516,
|
| 267 |
+
"learning_rate": 9.96237523753035e-05,
|
| 268 |
+
"loss": 0.0503,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.2695035460992908,
|
| 273 |
+
"grad_norm": 0.7772703766822815,
|
| 274 |
+
"learning_rate": 9.957436660008185e-05,
|
| 275 |
+
"loss": 0.0501,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.2765957446808511,
|
| 280 |
+
"grad_norm": 0.7801678776741028,
|
| 281 |
+
"learning_rate": 9.952195018143307e-05,
|
| 282 |
+
"loss": 0.0507,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.28368794326241137,
|
| 287 |
+
"grad_norm": 0.8867959976196289,
|
| 288 |
+
"learning_rate": 9.946650632374449e-05,
|
| 289 |
+
"loss": 0.0522,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.2907801418439716,
|
| 294 |
+
"grad_norm": 0.640887975692749,
|
| 295 |
+
"learning_rate": 9.940803841648078e-05,
|
| 296 |
+
"loss": 0.0437,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.2978723404255319,
|
| 301 |
+
"grad_norm": 0.9661104083061218,
|
| 302 |
+
"learning_rate": 9.934655003397662e-05,
|
| 303 |
+
"loss": 0.0485,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3049645390070922,
|
| 308 |
+
"grad_norm": 1.1752740144729614,
|
| 309 |
+
"learning_rate": 9.928204493521828e-05,
|
| 310 |
+
"loss": 0.0491,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.3120567375886525,
|
| 315 |
+
"grad_norm": 0.7718833684921265,
|
| 316 |
+
"learning_rate": 9.921452706361376e-05,
|
| 317 |
+
"loss": 0.0575,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3191489361702128,
|
| 322 |
+
"grad_norm": 0.7077618837356567,
|
| 323 |
+
"learning_rate": 9.91440005467517e-05,
|
| 324 |
+
"loss": 0.0466,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.3262411347517731,
|
| 329 |
+
"grad_norm": 0.7579907774925232,
|
| 330 |
+
"learning_rate": 9.907046969614912e-05,
|
| 331 |
+
"loss": 0.0511,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3333333333333333,
|
| 336 |
+
"grad_norm": 0.5514702796936035,
|
| 337 |
+
"learning_rate": 9.899393900698781e-05,
|
| 338 |
+
"loss": 0.047,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.3404255319148936,
|
| 343 |
+
"grad_norm": 0.8509230613708496,
|
| 344 |
+
"learning_rate": 9.891441315783948e-05,
|
| 345 |
+
"loss": 0.0428,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.3475177304964539,
|
| 350 |
+
"grad_norm": 0.46692895889282227,
|
| 351 |
+
"learning_rate": 9.883189701037981e-05,
|
| 352 |
+
"loss": 0.0456,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.3546099290780142,
|
| 357 |
+
"grad_norm": 0.5509600043296814,
|
| 358 |
+
"learning_rate": 9.874639560909117e-05,
|
| 359 |
+
"loss": 0.0437,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.3617021276595745,
|
| 364 |
+
"grad_norm": 0.5207144021987915,
|
| 365 |
+
"learning_rate": 9.865791418095437e-05,
|
| 366 |
+
"loss": 0.0436,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.36879432624113473,
|
| 371 |
+
"grad_norm": 0.5835493206977844,
|
| 372 |
+
"learning_rate": 9.856645813512892e-05,
|
| 373 |
+
"loss": 0.044,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.375886524822695,
|
| 378 |
+
"grad_norm": 1.0070799589157104,
|
| 379 |
+
"learning_rate": 9.847203306262249e-05,
|
| 380 |
+
"loss": 0.0461,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.3829787234042553,
|
| 385 |
+
"grad_norm": 1.2180109024047852,
|
| 386 |
+
"learning_rate": 9.837464473594911e-05,
|
| 387 |
+
"loss": 0.0552,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.3900709219858156,
|
| 392 |
+
"grad_norm": 0.6568053960800171,
|
| 393 |
+
"learning_rate": 9.827429910877619e-05,
|
| 394 |
+
"loss": 0.0516,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.3971631205673759,
|
| 399 |
+
"grad_norm": 1.0832271575927734,
|
| 400 |
+
"learning_rate": 9.817100231556065e-05,
|
| 401 |
+
"loss": 0.0485,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.40425531914893614,
|
| 406 |
+
"grad_norm": 0.8323943018913269,
|
| 407 |
+
"learning_rate": 9.806476067117384e-05,
|
| 408 |
+
"loss": 0.0439,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.41134751773049644,
|
| 413 |
+
"grad_norm": 0.5574596524238586,
|
| 414 |
+
"learning_rate": 9.79555806705155e-05,
|
| 415 |
+
"loss": 0.0405,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.41843971631205673,
|
| 420 |
+
"grad_norm": 0.5827070474624634,
|
| 421 |
+
"learning_rate": 9.784346898811674e-05,
|
| 422 |
+
"loss": 0.0446,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.425531914893617,
|
| 427 |
+
"grad_norm": 0.8284025192260742,
|
| 428 |
+
"learning_rate": 9.772843247773195e-05,
|
| 429 |
+
"loss": 0.0423,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4326241134751773,
|
| 434 |
+
"grad_norm": 0.5051993727684021,
|
| 435 |
+
"learning_rate": 9.761047817191986e-05,
|
| 436 |
+
"loss": 0.0401,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.4397163120567376,
|
| 441 |
+
"grad_norm": 0.7766433954238892,
|
| 442 |
+
"learning_rate": 9.748961328161356e-05,
|
| 443 |
+
"loss": 0.0441,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.44680851063829785,
|
| 448 |
+
"grad_norm": 0.48293229937553406,
|
| 449 |
+
"learning_rate": 9.736584519567975e-05,
|
| 450 |
+
"loss": 0.0455,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.45390070921985815,
|
| 455 |
+
"grad_norm": 0.7349538207054138,
|
| 456 |
+
"learning_rate": 9.723918148046696e-05,
|
| 457 |
+
"loss": 0.0383,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.46099290780141844,
|
| 462 |
+
"grad_norm": 0.7378718852996826,
|
| 463 |
+
"learning_rate": 9.710962987934304e-05,
|
| 464 |
+
"loss": 0.0425,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.46808510638297873,
|
| 469 |
+
"grad_norm": 0.5441968441009521,
|
| 470 |
+
"learning_rate": 9.697719831222172e-05,
|
| 471 |
+
"loss": 0.0411,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.475177304964539,
|
| 476 |
+
"grad_norm": 0.37726619839668274,
|
| 477 |
+
"learning_rate": 9.684189487507857e-05,
|
| 478 |
+
"loss": 0.0398,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.48226950354609927,
|
| 483 |
+
"grad_norm": 0.5768049359321594,
|
| 484 |
+
"learning_rate": 9.670372783945589e-05,
|
| 485 |
+
"loss": 0.0406,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.48936170212765956,
|
| 490 |
+
"grad_norm": 0.639546275138855,
|
| 491 |
+
"learning_rate": 9.656270565195718e-05,
|
| 492 |
+
"loss": 0.0377,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.49645390070921985,
|
| 497 |
+
"grad_norm": 0.5999423861503601,
|
| 498 |
+
"learning_rate": 9.641883693373076e-05,
|
| 499 |
+
"loss": 0.0421,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.5035460992907801,
|
| 504 |
+
"grad_norm": 0.46248650550842285,
|
| 505 |
+
"learning_rate": 9.627213047994264e-05,
|
| 506 |
+
"loss": 0.0337,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.5106382978723404,
|
| 511 |
+
"grad_norm": 0.36114415526390076,
|
| 512 |
+
"learning_rate": 9.612259525923893e-05,
|
| 513 |
+
"loss": 0.0327,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.5177304964539007,
|
| 518 |
+
"grad_norm": 0.49470576643943787,
|
| 519 |
+
"learning_rate": 9.59702404131975e-05,
|
| 520 |
+
"loss": 0.0371,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.524822695035461,
|
| 525 |
+
"grad_norm": 0.6341280341148376,
|
| 526 |
+
"learning_rate": 9.581507525576922e-05,
|
| 527 |
+
"loss": 0.0424,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.5319148936170213,
|
| 532 |
+
"grad_norm": 0.49816980957984924,
|
| 533 |
+
"learning_rate": 9.565710927270842e-05,
|
| 534 |
+
"loss": 0.0387,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.5390070921985816,
|
| 539 |
+
"grad_norm": 0.6306669116020203,
|
| 540 |
+
"learning_rate": 9.549635212099314e-05,
|
| 541 |
+
"loss": 0.039,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.5460992907801419,
|
| 546 |
+
"grad_norm": 0.691453218460083,
|
| 547 |
+
"learning_rate": 9.533281362823465e-05,
|
| 548 |
+
"loss": 0.0402,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.5531914893617021,
|
| 553 |
+
"grad_norm": 0.6569516062736511,
|
| 554 |
+
"learning_rate": 9.516650379207677e-05,
|
| 555 |
+
"loss": 0.0405,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.5602836879432624,
|
| 560 |
+
"grad_norm": 0.4964929223060608,
|
| 561 |
+
"learning_rate": 9.499743277958453e-05,
|
| 562 |
+
"loss": 0.0424,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.5673758865248227,
|
| 567 |
+
"grad_norm": 0.5208635330200195,
|
| 568 |
+
"learning_rate": 9.482561092662284e-05,
|
| 569 |
+
"loss": 0.039,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.574468085106383,
|
| 574 |
+
"grad_norm": 0.4303128719329834,
|
| 575 |
+
"learning_rate": 9.465104873722439e-05,
|
| 576 |
+
"loss": 0.0405,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.5815602836879432,
|
| 581 |
+
"grad_norm": 0.6891981959342957,
|
| 582 |
+
"learning_rate": 9.447375688294765e-05,
|
| 583 |
+
"loss": 0.0429,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.5886524822695035,
|
| 588 |
+
"grad_norm": 0.7455401420593262,
|
| 589 |
+
"learning_rate": 9.429374620222447e-05,
|
| 590 |
+
"loss": 0.0398,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.5957446808510638,
|
| 595 |
+
"grad_norm": 0.5415692925453186,
|
| 596 |
+
"learning_rate": 9.411102769969742e-05,
|
| 597 |
+
"loss": 0.0427,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.6028368794326241,
|
| 602 |
+
"grad_norm": 0.5630910396575928,
|
| 603 |
+
"learning_rate": 9.392561254554713e-05,
|
| 604 |
+
"loss": 0.0423,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.6099290780141844,
|
| 609 |
+
"grad_norm": 0.5906804800033569,
|
| 610 |
+
"learning_rate": 9.373751207480935e-05,
|
| 611 |
+
"loss": 0.043,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.6170212765957447,
|
| 616 |
+
"grad_norm": 0.7608093023300171,
|
| 617 |
+
"learning_rate": 9.354673778668206e-05,
|
| 618 |
+
"loss": 0.0416,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.624113475177305,
|
| 623 |
+
"grad_norm": 0.4081820547580719,
|
| 624 |
+
"learning_rate": 9.335330134382241e-05,
|
| 625 |
+
"loss": 0.0343,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.6312056737588653,
|
| 630 |
+
"grad_norm": 0.7236135601997375,
|
| 631 |
+
"learning_rate": 9.315721457163385e-05,
|
| 632 |
+
"loss": 0.0347,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.6382978723404256,
|
| 637 |
+
"grad_norm": 0.9877668619155884,
|
| 638 |
+
"learning_rate": 9.295848945754307e-05,
|
| 639 |
+
"loss": 0.0355,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.6453900709219859,
|
| 644 |
+
"grad_norm": 0.5710494518280029,
|
| 645 |
+
"learning_rate": 9.275713815026731e-05,
|
| 646 |
+
"loss": 0.0367,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.6524822695035462,
|
| 651 |
+
"grad_norm": 0.5722777247428894,
|
| 652 |
+
"learning_rate": 9.255317295907158e-05,
|
| 653 |
+
"loss": 0.0408,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.6595744680851063,
|
| 658 |
+
"grad_norm": 0.5658220052719116,
|
| 659 |
+
"learning_rate": 9.234660635301612e-05,
|
| 660 |
+
"loss": 0.0376,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.6666666666666666,
|
| 665 |
+
"grad_norm": 0.500984787940979,
|
| 666 |
+
"learning_rate": 9.213745096019432e-05,
|
| 667 |
+
"loss": 0.0356,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.6737588652482269,
|
| 672 |
+
"grad_norm": 0.6082512140274048,
|
| 673 |
+
"learning_rate": 9.192571956696043e-05,
|
| 674 |
+
"loss": 0.0399,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.6808510638297872,
|
| 679 |
+
"grad_norm": 0.4247320294380188,
|
| 680 |
+
"learning_rate": 9.171142511714814e-05,
|
| 681 |
+
"loss": 0.0368,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.6879432624113475,
|
| 686 |
+
"grad_norm": 0.5011168122291565,
|
| 687 |
+
"learning_rate": 9.149458071127914e-05,
|
| 688 |
+
"loss": 0.0388,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.6950354609929078,
|
| 693 |
+
"grad_norm": 0.60235595703125,
|
| 694 |
+
"learning_rate": 9.127519960576233e-05,
|
| 695 |
+
"loss": 0.0387,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.7021276595744681,
|
| 700 |
+
"grad_norm": 0.5439538359642029,
|
| 701 |
+
"learning_rate": 9.105329521208333e-05,
|
| 702 |
+
"loss": 0.0393,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.7092198581560284,
|
| 707 |
+
"grad_norm": 0.43570181727409363,
|
| 708 |
+
"learning_rate": 9.082888109598466e-05,
|
| 709 |
+
"loss": 0.04,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.7163120567375887,
|
| 714 |
+
"grad_norm": 0.6640782356262207,
|
| 715 |
+
"learning_rate": 9.060197097663634e-05,
|
| 716 |
+
"loss": 0.0407,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.723404255319149,
|
| 721 |
+
"grad_norm": 0.6727411150932312,
|
| 722 |
+
"learning_rate": 9.037257872579733e-05,
|
| 723 |
+
"loss": 0.039,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.7304964539007093,
|
| 728 |
+
"grad_norm": 0.45021480321884155,
|
| 729 |
+
"learning_rate": 9.014071836696735e-05,
|
| 730 |
+
"loss": 0.0339,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.7375886524822695,
|
| 735 |
+
"grad_norm": 0.5670303702354431,
|
| 736 |
+
"learning_rate": 8.990640407452966e-05,
|
| 737 |
+
"loss": 0.0348,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.7446808510638298,
|
| 742 |
+
"grad_norm": 0.44687455892562866,
|
| 743 |
+
"learning_rate": 8.966965017288455e-05,
|
| 744 |
+
"loss": 0.034,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.75177304964539,
|
| 749 |
+
"grad_norm": 0.4786987900733948,
|
| 750 |
+
"learning_rate": 8.943047113557358e-05,
|
| 751 |
+
"loss": 0.0316,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.7588652482269503,
|
| 756 |
+
"grad_norm": 0.5749778151512146,
|
| 757 |
+
"learning_rate": 8.91888815843948e-05,
|
| 758 |
+
"loss": 0.0344,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.7659574468085106,
|
| 763 |
+
"grad_norm": 0.7388485074043274,
|
| 764 |
+
"learning_rate": 8.89448962885089e-05,
|
| 765 |
+
"loss": 0.0348,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.7730496453900709,
|
| 770 |
+
"grad_norm": 0.6491293907165527,
|
| 771 |
+
"learning_rate": 8.869853016353626e-05,
|
| 772 |
+
"loss": 0.0318,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.7801418439716312,
|
| 777 |
+
"grad_norm": 0.5393919348716736,
|
| 778 |
+
"learning_rate": 8.844979827064516e-05,
|
| 779 |
+
"loss": 0.0346,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.7872340425531915,
|
| 784 |
+
"grad_norm": 0.5995272397994995,
|
| 785 |
+
"learning_rate": 8.819871581563098e-05,
|
| 786 |
+
"loss": 0.0347,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.7943262411347518,
|
| 791 |
+
"grad_norm": 0.647890567779541,
|
| 792 |
+
"learning_rate": 8.794529814798674e-05,
|
| 793 |
+
"loss": 0.0338,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.8014184397163121,
|
| 798 |
+
"grad_norm": 0.46107402443885803,
|
| 799 |
+
"learning_rate": 8.768956075996459e-05,
|
| 800 |
+
"loss": 0.0391,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.8085106382978723,
|
| 805 |
+
"grad_norm": 0.5663193464279175,
|
| 806 |
+
"learning_rate": 8.743151928562883e-05,
|
| 807 |
+
"loss": 0.0364,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.8156028368794326,
|
| 812 |
+
"grad_norm": 0.41998445987701416,
|
| 813 |
+
"learning_rate": 8.717118949990006e-05,
|
| 814 |
+
"loss": 0.0322,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.8226950354609929,
|
| 819 |
+
"grad_norm": 0.7538816332817078,
|
| 820 |
+
"learning_rate": 8.69085873175909e-05,
|
| 821 |
+
"loss": 0.0378,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.8297872340425532,
|
| 826 |
+
"grad_norm": 0.387187123298645,
|
| 827 |
+
"learning_rate": 8.664372879243297e-05,
|
| 828 |
+
"loss": 0.0369,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.8368794326241135,
|
| 833 |
+
"grad_norm": 0.5358660221099854,
|
| 834 |
+
"learning_rate": 8.637663011609556e-05,
|
| 835 |
+
"loss": 0.0355,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.8439716312056738,
|
| 840 |
+
"grad_norm": 0.4031761884689331,
|
| 841 |
+
"learning_rate": 8.610730761719572e-05,
|
| 842 |
+
"loss": 0.0376,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.851063829787234,
|
| 847 |
+
"grad_norm": 0.5619888305664062,
|
| 848 |
+
"learning_rate": 8.583577776030005e-05,
|
| 849 |
+
"loss": 0.0334,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.8581560283687943,
|
| 854 |
+
"grad_norm": 0.3582878112792969,
|
| 855 |
+
"learning_rate": 8.55620571449182e-05,
|
| 856 |
+
"loss": 0.0397,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.8652482269503546,
|
| 861 |
+
"grad_norm": 0.4317435622215271,
|
| 862 |
+
"learning_rate": 8.528616250448804e-05,
|
| 863 |
+
"loss": 0.0361,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.8723404255319149,
|
| 868 |
+
"grad_norm": 0.5297847986221313,
|
| 869 |
+
"learning_rate": 8.500811070535272e-05,
|
| 870 |
+
"loss": 0.028,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.8794326241134752,
|
| 875 |
+
"grad_norm": 0.5080351233482361,
|
| 876 |
+
"learning_rate": 8.472791874572959e-05,
|
| 877 |
+
"loss": 0.0328,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.8865248226950354,
|
| 882 |
+
"grad_norm": 0.5950552225112915,
|
| 883 |
+
"learning_rate": 8.444560375467097e-05,
|
| 884 |
+
"loss": 0.0384,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.8936170212765957,
|
| 889 |
+
"grad_norm": 0.5029882192611694,
|
| 890 |
+
"learning_rate": 8.41611829910171e-05,
|
| 891 |
+
"loss": 0.0359,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.900709219858156,
|
| 896 |
+
"grad_norm": 0.45763489603996277,
|
| 897 |
+
"learning_rate": 8.387467384234095e-05,
|
| 898 |
+
"loss": 0.0322,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.9078014184397163,
|
| 903 |
+
"grad_norm": 0.6697009801864624,
|
| 904 |
+
"learning_rate": 8.358609382388538e-05,
|
| 905 |
+
"loss": 0.0315,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.9148936170212766,
|
| 910 |
+
"grad_norm": 0.6710605025291443,
|
| 911 |
+
"learning_rate": 8.32954605774922e-05,
|
| 912 |
+
"loss": 0.0407,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.9219858156028369,
|
| 917 |
+
"grad_norm": 0.3776458203792572,
|
| 918 |
+
"learning_rate": 8.300279187052387e-05,
|
| 919 |
+
"loss": 0.0362,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9290780141843972,
|
| 924 |
+
"grad_norm": 0.5019909143447876,
|
| 925 |
+
"learning_rate": 8.270810559477715e-05,
|
| 926 |
+
"loss": 0.032,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.9361702127659575,
|
| 931 |
+
"grad_norm": 0.49398350715637207,
|
| 932 |
+
"learning_rate": 8.241141976538943e-05,
|
| 933 |
+
"loss": 0.0358,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.9432624113475178,
|
| 938 |
+
"grad_norm": 0.3629569709300995,
|
| 939 |
+
"learning_rate": 8.211275251973734e-05,
|
| 940 |
+
"loss": 0.0388,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.950354609929078,
|
| 945 |
+
"grad_norm": 0.2770695388317108,
|
| 946 |
+
"learning_rate": 8.181212211632799e-05,
|
| 947 |
+
"loss": 0.0389,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.9574468085106383,
|
| 952 |
+
"grad_norm": 0.5267712473869324,
|
| 953 |
+
"learning_rate": 8.150954693368277e-05,
|
| 954 |
+
"loss": 0.0338,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.9645390070921985,
|
| 959 |
+
"grad_norm": 0.6642586588859558,
|
| 960 |
+
"learning_rate": 8.120504546921375e-05,
|
| 961 |
+
"loss": 0.038,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.9716312056737588,
|
| 966 |
+
"grad_norm": 0.47881484031677246,
|
| 967 |
+
"learning_rate": 8.089863633809299e-05,
|
| 968 |
+
"loss": 0.0287,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.9787234042553191,
|
| 973 |
+
"grad_norm": 0.5986367464065552,
|
| 974 |
+
"learning_rate": 8.059033827211438e-05,
|
| 975 |
+
"loss": 0.0334,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.9858156028368794,
|
| 980 |
+
"grad_norm": 0.5382274985313416,
|
| 981 |
+
"learning_rate": 8.028017011854861e-05,
|
| 982 |
+
"loss": 0.0355,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.9929078014184397,
|
| 987 |
+
"grad_norm": 0.427786260843277,
|
| 988 |
+
"learning_rate": 7.9968150838991e-05,
|
| 989 |
+
"loss": 0.0395,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.0,
|
| 994 |
+
"grad_norm": 0.8537722826004028,
|
| 995 |
+
"learning_rate": 7.965429950820222e-05,
|
| 996 |
+
"loss": 0.0338,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.0070921985815602,
|
| 1001 |
+
"grad_norm": 0.4295312464237213,
|
| 1002 |
+
"learning_rate": 7.933863531294225e-05,
|
| 1003 |
+
"loss": 0.0285,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.0141843971631206,
|
| 1008 |
+
"grad_norm": 0.4605224132537842,
|
| 1009 |
+
"learning_rate": 7.902117755079738e-05,
|
| 1010 |
+
"loss": 0.0354,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.0212765957446808,
|
| 1015 |
+
"grad_norm": 0.3009595572948456,
|
| 1016 |
+
"learning_rate": 7.870194562900055e-05,
|
| 1017 |
+
"loss": 0.0272,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.0283687943262412,
|
| 1022 |
+
"grad_norm": 0.5193508267402649,
|
| 1023 |
+
"learning_rate": 7.838095906324492e-05,
|
| 1024 |
+
"loss": 0.0332,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.0354609929078014,
|
| 1029 |
+
"grad_norm": 0.530876100063324,
|
| 1030 |
+
"learning_rate": 7.805823747649072e-05,
|
| 1031 |
+
"loss": 0.0347,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.0425531914893618,
|
| 1036 |
+
"grad_norm": 0.5352266430854797,
|
| 1037 |
+
"learning_rate": 7.773380059776574e-05,
|
| 1038 |
+
"loss": 0.0287,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.049645390070922,
|
| 1043 |
+
"grad_norm": 0.6278623342514038,
|
| 1044 |
+
"learning_rate": 7.740766826095918e-05,
|
| 1045 |
+
"loss": 0.0317,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.0567375886524824,
|
| 1050 |
+
"grad_norm": 0.6249827742576599,
|
| 1051 |
+
"learning_rate": 7.70798604036091e-05,
|
| 1052 |
+
"loss": 0.0296,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.0638297872340425,
|
| 1057 |
+
"grad_norm": 0.31095314025878906,
|
| 1058 |
+
"learning_rate": 7.675039706568373e-05,
|
| 1059 |
+
"loss": 0.0297,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.070921985815603,
|
| 1064 |
+
"grad_norm": 0.5677725672721863,
|
| 1065 |
+
"learning_rate": 7.641929838835613e-05,
|
| 1066 |
+
"loss": 0.0334,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.0780141843971631,
|
| 1071 |
+
"grad_norm": 0.607011079788208,
|
| 1072 |
+
"learning_rate": 7.608658461277304e-05,
|
| 1073 |
+
"loss": 0.0332,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.0851063829787233,
|
| 1078 |
+
"grad_norm": 0.4887952506542206,
|
| 1079 |
+
"learning_rate": 7.575227607881757e-05,
|
| 1080 |
+
"loss": 0.0316,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.0921985815602837,
|
| 1085 |
+
"grad_norm": 0.41991710662841797,
|
| 1086 |
+
"learning_rate": 7.541639322386545e-05,
|
| 1087 |
+
"loss": 0.0334,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.099290780141844,
|
| 1092 |
+
"grad_norm": 0.3826978802680969,
|
| 1093 |
+
"learning_rate": 7.507895658153594e-05,
|
| 1094 |
+
"loss": 0.0282,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.1063829787234043,
|
| 1099 |
+
"grad_norm": 0.40258097648620605,
|
| 1100 |
+
"learning_rate": 7.473998678043634e-05,
|
| 1101 |
+
"loss": 0.0266,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.1134751773049645,
|
| 1106 |
+
"grad_norm": 0.5759483575820923,
|
| 1107 |
+
"learning_rate": 7.439950454290102e-05,
|
| 1108 |
+
"loss": 0.0322,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.1205673758865249,
|
| 1113 |
+
"grad_norm": 0.44756197929382324,
|
| 1114 |
+
"learning_rate": 7.40575306837245e-05,
|
| 1115 |
+
"loss": 0.0284,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.127659574468085,
|
| 1120 |
+
"grad_norm": 0.7017154693603516,
|
| 1121 |
+
"learning_rate": 7.371408610888907e-05,
|
| 1122 |
+
"loss": 0.0296,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.1347517730496455,
|
| 1127 |
+
"grad_norm": 0.39028796553611755,
|
| 1128 |
+
"learning_rate": 7.33691918142866e-05,
|
| 1129 |
+
"loss": 0.029,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.1418439716312057,
|
| 1134 |
+
"grad_norm": 0.49805375933647156,
|
| 1135 |
+
"learning_rate": 7.30228688844352e-05,
|
| 1136 |
+
"loss": 0.0336,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.148936170212766,
|
| 1141 |
+
"grad_norm": 0.5060091614723206,
|
| 1142 |
+
"learning_rate": 7.267513849119001e-05,
|
| 1143 |
+
"loss": 0.0266,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.1560283687943262,
|
| 1148 |
+
"grad_norm": 0.44427555799484253,
|
| 1149 |
+
"learning_rate": 7.23260218924491e-05,
|
| 1150 |
+
"loss": 0.0266,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.1631205673758864,
|
| 1155 |
+
"grad_norm": 0.45797792077064514,
|
| 1156 |
+
"learning_rate": 7.197554043085378e-05,
|
| 1157 |
+
"loss": 0.0355,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.1702127659574468,
|
| 1162 |
+
"grad_norm": 0.4404367506504059,
|
| 1163 |
+
"learning_rate": 7.162371553248392e-05,
|
| 1164 |
+
"loss": 0.0276,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.177304964539007,
|
| 1169 |
+
"grad_norm": 0.5416800379753113,
|
| 1170 |
+
"learning_rate": 7.127056870554807e-05,
|
| 1171 |
+
"loss": 0.0326,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.1843971631205674,
|
| 1176 |
+
"grad_norm": 0.4279535710811615,
|
| 1177 |
+
"learning_rate": 7.091612153906864e-05,
|
| 1178 |
+
"loss": 0.034,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.1914893617021276,
|
| 1183 |
+
"grad_norm": 0.6757423281669617,
|
| 1184 |
+
"learning_rate": 7.056039570156196e-05,
|
| 1185 |
+
"loss": 0.0325,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.198581560283688,
|
| 1190 |
+
"grad_norm": 0.5656224489212036,
|
| 1191 |
+
"learning_rate": 7.020341293971381e-05,
|
| 1192 |
+
"loss": 0.0319,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.2056737588652482,
|
| 1197 |
+
"grad_norm": 0.6633424162864685,
|
| 1198 |
+
"learning_rate": 6.984519507704986e-05,
|
| 1199 |
+
"loss": 0.0318,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.2127659574468086,
|
| 1204 |
+
"grad_norm": 0.47362446784973145,
|
| 1205 |
+
"learning_rate": 6.948576401260146e-05,
|
| 1206 |
+
"loss": 0.0305,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.2198581560283688,
|
| 1211 |
+
"grad_norm": 0.5503174066543579,
|
| 1212 |
+
"learning_rate": 6.912514171956704e-05,
|
| 1213 |
+
"loss": 0.0264,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.226950354609929,
|
| 1218 |
+
"grad_norm": 0.47481873631477356,
|
| 1219 |
+
"learning_rate": 6.876335024396872e-05,
|
| 1220 |
+
"loss": 0.0284,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.2340425531914894,
|
| 1225 |
+
"grad_norm": 0.43006354570388794,
|
| 1226 |
+
"learning_rate": 6.840041170330454e-05,
|
| 1227 |
+
"loss": 0.0388,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.2411347517730495,
|
| 1232 |
+
"grad_norm": 0.6476072669029236,
|
| 1233 |
+
"learning_rate": 6.803634828519643e-05,
|
| 1234 |
+
"loss": 0.0304,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.24822695035461,
|
| 1239 |
+
"grad_norm": 0.613460898399353,
|
| 1240 |
+
"learning_rate": 6.767118224603373e-05,
|
| 1241 |
+
"loss": 0.0274,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.2553191489361701,
|
| 1246 |
+
"grad_norm": 0.35610130429267883,
|
| 1247 |
+
"learning_rate": 6.730493590961267e-05,
|
| 1248 |
+
"loss": 0.0345,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.2624113475177305,
|
| 1253 |
+
"grad_norm": 0.40915030241012573,
|
| 1254 |
+
"learning_rate": 6.693763166577154e-05,
|
| 1255 |
+
"loss": 0.0298,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.2695035460992907,
|
| 1260 |
+
"grad_norm": 0.2953985035419464,
|
| 1261 |
+
"learning_rate": 6.656929196902195e-05,
|
| 1262 |
+
"loss": 0.0274,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.2765957446808511,
|
| 1267 |
+
"grad_norm": 0.4407590925693512,
|
| 1268 |
+
"learning_rate": 6.619993933717623e-05,
|
| 1269 |
+
"loss": 0.029,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.2836879432624113,
|
| 1274 |
+
"grad_norm": 0.3795197606086731,
|
| 1275 |
+
"learning_rate": 6.582959634997068e-05,
|
| 1276 |
+
"loss": 0.0251,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.2907801418439715,
|
| 1281 |
+
"grad_norm": 0.33862507343292236,
|
| 1282 |
+
"learning_rate": 6.54582856476853e-05,
|
| 1283 |
+
"loss": 0.0296,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.297872340425532,
|
| 1288 |
+
"grad_norm": 0.38457411527633667,
|
| 1289 |
+
"learning_rate": 6.508602992975963e-05,
|
| 1290 |
+
"loss": 0.0249,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.3049645390070923,
|
| 1295 |
+
"grad_norm": 0.5313107967376709,
|
| 1296 |
+
"learning_rate": 6.471285195340518e-05,
|
| 1297 |
+
"loss": 0.0262,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.3120567375886525,
|
| 1302 |
+
"grad_norm": 0.44097694754600525,
|
| 1303 |
+
"learning_rate": 6.43387745322141e-05,
|
| 1304 |
+
"loss": 0.0291,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.3191489361702127,
|
| 1309 |
+
"grad_norm": 0.5393781661987305,
|
| 1310 |
+
"learning_rate": 6.396382053476459e-05,
|
| 1311 |
+
"loss": 0.0284,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.326241134751773,
|
| 1316 |
+
"grad_norm": 0.2834714949131012,
|
| 1317 |
+
"learning_rate": 6.358801288322274e-05,
|
| 1318 |
+
"loss": 0.0275,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.3333333333333333,
|
| 1323 |
+
"grad_norm": 0.46584266424179077,
|
| 1324 |
+
"learning_rate": 6.321137455194141e-05,
|
| 1325 |
+
"loss": 0.0319,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.3404255319148937,
|
| 1330 |
+
"grad_norm": 0.4538133442401886,
|
| 1331 |
+
"learning_rate": 6.283392856605559e-05,
|
| 1332 |
+
"loss": 0.0265,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.3475177304964538,
|
| 1337 |
+
"grad_norm": 0.4815518260002136,
|
| 1338 |
+
"learning_rate": 6.245569800007484e-05,
|
| 1339 |
+
"loss": 0.0331,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.3546099290780143,
|
| 1344 |
+
"grad_norm": 0.5020127892494202,
|
| 1345 |
+
"learning_rate": 6.207670597647265e-05,
|
| 1346 |
+
"loss": 0.0277,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.3617021276595744,
|
| 1351 |
+
"grad_norm": 0.5370143055915833,
|
| 1352 |
+
"learning_rate": 6.169697566427299e-05,
|
| 1353 |
+
"loss": 0.027,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.3687943262411348,
|
| 1358 |
+
"grad_norm": 0.5647911429405212,
|
| 1359 |
+
"learning_rate": 6.131653027763373e-05,
|
| 1360 |
+
"loss": 0.03,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.375886524822695,
|
| 1365 |
+
"grad_norm": 0.44125860929489136,
|
| 1366 |
+
"learning_rate": 6.093539307442761e-05,
|
| 1367 |
+
"loss": 0.0323,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.3829787234042552,
|
| 1372 |
+
"grad_norm": 0.5117540955543518,
|
| 1373 |
+
"learning_rate": 6.055358735482045e-05,
|
| 1374 |
+
"loss": 0.0305,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.3900709219858156,
|
| 1379 |
+
"grad_norm": 0.3357088565826416,
|
| 1380 |
+
"learning_rate": 6.01711364598466e-05,
|
| 1381 |
+
"loss": 0.0279,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.397163120567376,
|
| 1386 |
+
"grad_norm": 0.3001289665699005,
|
| 1387 |
+
"learning_rate": 5.97880637699821e-05,
|
| 1388 |
+
"loss": 0.0252,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.4042553191489362,
|
| 1393 |
+
"grad_norm": 0.7187599539756775,
|
| 1394 |
+
"learning_rate": 5.9404392703715374e-05,
|
| 1395 |
+
"loss": 0.0266,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.4113475177304964,
|
| 1400 |
+
"grad_norm": 0.6726476550102234,
|
| 1401 |
+
"learning_rate": 5.902014671611552e-05,
|
| 1402 |
+
"loss": 0.0256,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.4184397163120568,
|
| 1407 |
+
"grad_norm": 0.5328847765922546,
|
| 1408 |
+
"learning_rate": 5.863534929739851e-05,
|
| 1409 |
+
"loss": 0.0347,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.425531914893617,
|
| 1414 |
+
"grad_norm": 0.5943588614463806,
|
| 1415 |
+
"learning_rate": 5.825002397149105e-05,
|
| 1416 |
+
"loss": 0.0317,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.4326241134751774,
|
| 1421 |
+
"grad_norm": 0.5321977138519287,
|
| 1422 |
+
"learning_rate": 5.786419429459261e-05,
|
| 1423 |
+
"loss": 0.0258,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.4397163120567376,
|
| 1428 |
+
"grad_norm": 0.5272582173347473,
|
| 1429 |
+
"learning_rate": 5.7477883853735215e-05,
|
| 1430 |
+
"loss": 0.028,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.4468085106382977,
|
| 1435 |
+
"grad_norm": 0.3440077006816864,
|
| 1436 |
+
"learning_rate": 5.7091116265341605e-05,
|
| 1437 |
+
"loss": 0.0305,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.4539007092198581,
|
| 1442 |
+
"grad_norm": 0.3508741557598114,
|
| 1443 |
+
"learning_rate": 5.670391517378144e-05,
|
| 1444 |
+
"loss": 0.0309,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.4609929078014185,
|
| 1449 |
+
"grad_norm": 0.3721978962421417,
|
| 1450 |
+
"learning_rate": 5.631630424992588e-05,
|
| 1451 |
+
"loss": 0.023,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.4680851063829787,
|
| 1456 |
+
"grad_norm": 0.6611557006835938,
|
| 1457 |
+
"learning_rate": 5.592830718970041e-05,
|
| 1458 |
+
"loss": 0.0263,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.475177304964539,
|
| 1463 |
+
"grad_norm": 0.6382917761802673,
|
| 1464 |
+
"learning_rate": 5.553994771263633e-05,
|
| 1465 |
+
"loss": 0.0322,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.4822695035460993,
|
| 1470 |
+
"grad_norm": 0.45870348811149597,
|
| 1471 |
+
"learning_rate": 5.515124956042069e-05,
|
| 1472 |
+
"loss": 0.0286,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.4893617021276595,
|
| 1477 |
+
"grad_norm": 0.34599244594573975,
|
| 1478 |
+
"learning_rate": 5.476223649544485e-05,
|
| 1479 |
+
"loss": 0.0303,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.49645390070922,
|
| 1484 |
+
"grad_norm": 0.5597394704818726,
|
| 1485 |
+
"learning_rate": 5.4372932299351774e-05,
|
| 1486 |
+
"loss": 0.0261,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.50354609929078,
|
| 1491 |
+
"grad_norm": 0.34684792160987854,
|
| 1492 |
+
"learning_rate": 5.39833607715823e-05,
|
| 1493 |
+
"loss": 0.0245,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.5106382978723403,
|
| 1498 |
+
"grad_norm": 0.43969881534576416,
|
| 1499 |
+
"learning_rate": 5.359354572792006e-05,
|
| 1500 |
+
"loss": 0.0312,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.5177304964539007,
|
| 1505 |
+
"grad_norm": 0.399265855550766,
|
| 1506 |
+
"learning_rate": 5.320351099903565e-05,
|
| 1507 |
+
"loss": 0.0285,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.524822695035461,
|
| 1512 |
+
"grad_norm": 0.37245163321495056,
|
| 1513 |
+
"learning_rate": 5.281328042902973e-05,
|
| 1514 |
+
"loss": 0.0304,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.5319148936170213,
|
| 1519 |
+
"grad_norm": 0.5175348520278931,
|
| 1520 |
+
"learning_rate": 5.2422877873975384e-05,
|
| 1521 |
+
"loss": 0.0225,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.5390070921985815,
|
| 1526 |
+
"grad_norm": 0.435390442609787,
|
| 1527 |
+
"learning_rate": 5.2032327200459665e-05,
|
| 1528 |
+
"loss": 0.0246,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.5460992907801419,
|
| 1533 |
+
"grad_norm": 0.384159117937088,
|
| 1534 |
+
"learning_rate": 5.1641652284124645e-05,
|
| 1535 |
+
"loss": 0.0259,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.5531914893617023,
|
| 1540 |
+
"grad_norm": 0.26652634143829346,
|
| 1541 |
+
"learning_rate": 5.125087700820772e-05,
|
| 1542 |
+
"loss": 0.0264,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.5602836879432624,
|
| 1547 |
+
"grad_norm": 0.28009992837905884,
|
| 1548 |
+
"learning_rate": 5.0860025262081656e-05,
|
| 1549 |
+
"loss": 0.0279,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.5673758865248226,
|
| 1554 |
+
"grad_norm": 0.43979257345199585,
|
| 1555 |
+
"learning_rate": 5.0469120939794015e-05,
|
| 1556 |
+
"loss": 0.0296,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.574468085106383,
|
| 1561 |
+
"grad_norm": 0.5793026089668274,
|
| 1562 |
+
"learning_rate": 5.0078187938606556e-05,
|
| 1563 |
+
"loss": 0.0309,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.5815602836879432,
|
| 1568 |
+
"grad_norm": 0.4140380024909973,
|
| 1569 |
+
"learning_rate": 4.968725015753429e-05,
|
| 1570 |
+
"loss": 0.0292,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.5886524822695036,
|
| 1575 |
+
"grad_norm": 0.2710001468658447,
|
| 1576 |
+
"learning_rate": 4.92963314958844e-05,
|
| 1577 |
+
"loss": 0.0232,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.5957446808510638,
|
| 1582 |
+
"grad_norm": 0.5336018800735474,
|
| 1583 |
+
"learning_rate": 4.890545585179527e-05,
|
| 1584 |
+
"loss": 0.0261,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.602836879432624,
|
| 1589 |
+
"grad_norm": 0.4774004817008972,
|
| 1590 |
+
"learning_rate": 4.851464712077546e-05,
|
| 1591 |
+
"loss": 0.0248,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.6099290780141844,
|
| 1596 |
+
"grad_norm": 0.30671364068984985,
|
| 1597 |
+
"learning_rate": 4.81239291942429e-05,
|
| 1598 |
+
"loss": 0.0227,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.6170212765957448,
|
| 1603 |
+
"grad_norm": 0.36705008149147034,
|
| 1604 |
+
"learning_rate": 4.7733325958064365e-05,
|
| 1605 |
+
"loss": 0.0288,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.624113475177305,
|
| 1610 |
+
"grad_norm": 0.6876405477523804,
|
| 1611 |
+
"learning_rate": 4.734286129109524e-05,
|
| 1612 |
+
"loss": 0.0298,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.6312056737588652,
|
| 1617 |
+
"grad_norm": 0.40389272570610046,
|
| 1618 |
+
"learning_rate": 4.6952559063719744e-05,
|
| 1619 |
+
"loss": 0.0284,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.6382978723404256,
|
| 1624 |
+
"grad_norm": 0.23327293992042542,
|
| 1625 |
+
"learning_rate": 4.656244313639153e-05,
|
| 1626 |
+
"loss": 0.0266,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.645390070921986,
|
| 1631 |
+
"grad_norm": 0.33936405181884766,
|
| 1632 |
+
"learning_rate": 4.6172537358175214e-05,
|
| 1633 |
+
"loss": 0.0271,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.6524822695035462,
|
| 1638 |
+
"grad_norm": 0.7294065952301025,
|
| 1639 |
+
"learning_rate": 4.57828655652883e-05,
|
| 1640 |
+
"loss": 0.029,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.6595744680851063,
|
| 1645 |
+
"grad_norm": 0.39774230122566223,
|
| 1646 |
+
"learning_rate": 4.539345157964399e-05,
|
| 1647 |
+
"loss": 0.0324,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.6666666666666665,
|
| 1652 |
+
"grad_norm": 0.5822659134864807,
|
| 1653 |
+
"learning_rate": 4.5004319207394815e-05,
|
| 1654 |
+
"loss": 0.0291,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.673758865248227,
|
| 1659 |
+
"grad_norm": 0.43963325023651123,
|
| 1660 |
+
"learning_rate": 4.461549223747749e-05,
|
| 1661 |
+
"loss": 0.0216,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.6808510638297873,
|
| 1666 |
+
"grad_norm": 0.4312455356121063,
|
| 1667 |
+
"learning_rate": 4.422699444015838e-05,
|
| 1668 |
+
"loss": 0.0282,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.6879432624113475,
|
| 1673 |
+
"grad_norm": 0.2643612027168274,
|
| 1674 |
+
"learning_rate": 4.383884956558051e-05,
|
| 1675 |
+
"loss": 0.024,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.6950354609929077,
|
| 1680 |
+
"grad_norm": 0.33632996678352356,
|
| 1681 |
+
"learning_rate": 4.3451081342311516e-05,
|
| 1682 |
+
"loss": 0.0274,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.702127659574468,
|
| 1687 |
+
"grad_norm": 0.25713565945625305,
|
| 1688 |
+
"learning_rate": 4.306371347589318e-05,
|
| 1689 |
+
"loss": 0.0266,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.7092198581560285,
|
| 1694 |
+
"grad_norm": 0.44971826672554016,
|
| 1695 |
+
"learning_rate": 4.267676964739211e-05,
|
| 1696 |
+
"loss": 0.0274,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.7163120567375887,
|
| 1701 |
+
"grad_norm": 0.4082903265953064,
|
| 1702 |
+
"learning_rate": 4.2290273511952124e-05,
|
| 1703 |
+
"loss": 0.0252,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.7234042553191489,
|
| 1708 |
+
"grad_norm": 0.43493014574050903,
|
| 1709 |
+
"learning_rate": 4.190424869734805e-05,
|
| 1710 |
+
"loss": 0.0287,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.7304964539007093,
|
| 1715 |
+
"grad_norm": 0.5412095189094543,
|
| 1716 |
+
"learning_rate": 4.1518718802541365e-05,
|
| 1717 |
+
"loss": 0.0247,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.7375886524822695,
|
| 1722 |
+
"grad_norm": 0.541191816329956,
|
| 1723 |
+
"learning_rate": 4.113370739623746e-05,
|
| 1724 |
+
"loss": 0.0249,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.7446808510638299,
|
| 1729 |
+
"grad_norm": 0.5570827722549438,
|
| 1730 |
+
"learning_rate": 4.074923801544484e-05,
|
| 1731 |
+
"loss": 0.024,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.75177304964539,
|
| 1736 |
+
"grad_norm": 0.37162888050079346,
|
| 1737 |
+
"learning_rate": 4.036533416403621e-05,
|
| 1738 |
+
"loss": 0.0314,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.7588652482269502,
|
| 1743 |
+
"grad_norm": 0.626410722732544,
|
| 1744 |
+
"learning_rate": 3.9982019311311594e-05,
|
| 1745 |
+
"loss": 0.0239,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.7659574468085106,
|
| 1750 |
+
"grad_norm": 0.4442892372608185,
|
| 1751 |
+
"learning_rate": 3.959931689056361e-05,
|
| 1752 |
+
"loss": 0.0227,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.773049645390071,
|
| 1757 |
+
"grad_norm": 0.5664969086647034,
|
| 1758 |
+
"learning_rate": 3.9217250297644936e-05,
|
| 1759 |
+
"loss": 0.0323,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.7801418439716312,
|
| 1764 |
+
"grad_norm": 0.4956495761871338,
|
| 1765 |
+
"learning_rate": 3.8835842889537935e-05,
|
| 1766 |
+
"loss": 0.0235,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.7872340425531914,
|
| 1771 |
+
"grad_norm": 0.5894970893859863,
|
| 1772 |
+
"learning_rate": 3.845511798292692e-05,
|
| 1773 |
+
"loss": 0.0261,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.7943262411347518,
|
| 1778 |
+
"grad_norm": 0.3302382230758667,
|
| 1779 |
+
"learning_rate": 3.8075098852772604e-05,
|
| 1780 |
+
"loss": 0.0226,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.8014184397163122,
|
| 1785 |
+
"grad_norm": 0.4544001817703247,
|
| 1786 |
+
"learning_rate": 3.76958087308893e-05,
|
| 1787 |
+
"loss": 0.0234,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.8085106382978724,
|
| 1792 |
+
"grad_norm": 0.6900862455368042,
|
| 1793 |
+
"learning_rate": 3.731727080452464e-05,
|
| 1794 |
+
"loss": 0.0279,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.8156028368794326,
|
| 1799 |
+
"grad_norm": 0.5951893329620361,
|
| 1800 |
+
"learning_rate": 3.693950821494209e-05,
|
| 1801 |
+
"loss": 0.024,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.8226950354609928,
|
| 1806 |
+
"grad_norm": 0.26025912165641785,
|
| 1807 |
+
"learning_rate": 3.6562544056006206e-05,
|
| 1808 |
+
"loss": 0.024,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.8297872340425532,
|
| 1813 |
+
"grad_norm": 0.38223400712013245,
|
| 1814 |
+
"learning_rate": 3.6186401372770966e-05,
|
| 1815 |
+
"loss": 0.0269,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.8368794326241136,
|
| 1820 |
+
"grad_norm": 0.3878083825111389,
|
| 1821 |
+
"learning_rate": 3.5811103160070696e-05,
|
| 1822 |
+
"loss": 0.0271,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.8439716312056738,
|
| 1827 |
+
"grad_norm": 0.5003122687339783,
|
| 1828 |
+
"learning_rate": 3.5436672361114576e-05,
|
| 1829 |
+
"loss": 0.0256,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.851063829787234,
|
| 1834 |
+
"grad_norm": 0.3435218632221222,
|
| 1835 |
+
"learning_rate": 3.506313186608392e-05,
|
| 1836 |
+
"loss": 0.0209,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.8581560283687943,
|
| 1841 |
+
"grad_norm": 0.3536638021469116,
|
| 1842 |
+
"learning_rate": 3.469050451073287e-05,
|
| 1843 |
+
"loss": 0.0227,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.8652482269503547,
|
| 1848 |
+
"grad_norm": 0.2422986626625061,
|
| 1849 |
+
"learning_rate": 3.431881307499225e-05,
|
| 1850 |
+
"loss": 0.0212,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.872340425531915,
|
| 1855 |
+
"grad_norm": 0.3235369026660919,
|
| 1856 |
+
"learning_rate": 3.394808028157718e-05,
|
| 1857 |
+
"loss": 0.026,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.8794326241134751,
|
| 1862 |
+
"grad_norm": 0.32172736525535583,
|
| 1863 |
+
"learning_rate": 3.357832879459776e-05,
|
| 1864 |
+
"loss": 0.0226,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.8865248226950353,
|
| 1869 |
+
"grad_norm": 0.3710137605667114,
|
| 1870 |
+
"learning_rate": 3.320958121817363e-05,
|
| 1871 |
+
"loss": 0.0261,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.8936170212765957,
|
| 1876 |
+
"grad_norm": 0.2597503662109375,
|
| 1877 |
+
"learning_rate": 3.28418600950521e-05,
|
| 1878 |
+
"loss": 0.0226,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.900709219858156,
|
| 1883 |
+
"grad_norm": 0.4243910014629364,
|
| 1884 |
+
"learning_rate": 3.247518790522999e-05,
|
| 1885 |
+
"loss": 0.0272,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.9078014184397163,
|
| 1890 |
+
"grad_norm": 0.35146793723106384,
|
| 1891 |
+
"learning_rate": 3.210958706457944e-05,
|
| 1892 |
+
"loss": 0.0236,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.9148936170212765,
|
| 1897 |
+
"grad_norm": 0.5099648833274841,
|
| 1898 |
+
"learning_rate": 3.174507992347746e-05,
|
| 1899 |
+
"loss": 0.0231,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.9219858156028369,
|
| 1904 |
+
"grad_norm": 0.3687443435192108,
|
| 1905 |
+
"learning_rate": 3.1381688765439685e-05,
|
| 1906 |
+
"loss": 0.0226,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.9290780141843973,
|
| 1911 |
+
"grad_norm": 0.39274775981903076,
|
| 1912 |
+
"learning_rate": 3.101943580575798e-05,
|
| 1913 |
+
"loss": 0.0203,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.9361702127659575,
|
| 1918 |
+
"grad_norm": 0.38869839906692505,
|
| 1919 |
+
"learning_rate": 3.065834319014245e-05,
|
| 1920 |
+
"loss": 0.0259,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.9432624113475176,
|
| 1925 |
+
"grad_norm": 0.3917119801044464,
|
| 1926 |
+
"learning_rate": 3.0298432993367574e-05,
|
| 1927 |
+
"loss": 0.0272,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.950354609929078,
|
| 1932 |
+
"grad_norm": 0.3306979835033417,
|
| 1933 |
+
"learning_rate": 2.9939727217922686e-05,
|
| 1934 |
+
"loss": 0.0213,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.9574468085106385,
|
| 1939 |
+
"grad_norm": 0.3652242422103882,
|
| 1940 |
+
"learning_rate": 2.9582247792666874e-05,
|
| 1941 |
+
"loss": 0.0199,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.9645390070921986,
|
| 1946 |
+
"grad_norm": 0.27394425868988037,
|
| 1947 |
+
"learning_rate": 2.922601657148847e-05,
|
| 1948 |
+
"loss": 0.0242,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.9716312056737588,
|
| 1953 |
+
"grad_norm": 0.341784805059433,
|
| 1954 |
+
"learning_rate": 2.887105533196895e-05,
|
| 1955 |
+
"loss": 0.0278,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.978723404255319,
|
| 1960 |
+
"grad_norm": 0.320995569229126,
|
| 1961 |
+
"learning_rate": 2.8517385774051695e-05,
|
| 1962 |
+
"loss": 0.0214,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.9858156028368794,
|
| 1967 |
+
"grad_norm": 0.6058695316314697,
|
| 1968 |
+
"learning_rate": 2.8165029518715335e-05,
|
| 1969 |
+
"loss": 0.0218,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.9929078014184398,
|
| 1974 |
+
"grad_norm": 0.20845060050487518,
|
| 1975 |
+
"learning_rate": 2.7814008106652012e-05,
|
| 1976 |
+
"loss": 0.0216,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.0,
|
| 1981 |
+
"grad_norm": 0.6698796153068542,
|
| 1982 |
+
"learning_rate": 2.7464342996950537e-05,
|
| 1983 |
+
"loss": 0.0239,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.00709219858156,
|
| 1988 |
+
"grad_norm": 0.3951570689678192,
|
| 1989 |
+
"learning_rate": 2.711605556578452e-05,
|
| 1990 |
+
"loss": 0.0181,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.0141843971631204,
|
| 1995 |
+
"grad_norm": 0.28108590841293335,
|
| 1996 |
+
"learning_rate": 2.676916710510552e-05,
|
| 1997 |
+
"loss": 0.0189,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.021276595744681,
|
| 2002 |
+
"grad_norm": 0.3553146421909332,
|
| 2003 |
+
"learning_rate": 2.6423698821341507e-05,
|
| 2004 |
+
"loss": 0.023,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.028368794326241,
|
| 2009 |
+
"grad_norm": 0.354978084564209,
|
| 2010 |
+
"learning_rate": 2.6079671834100354e-05,
|
| 2011 |
+
"loss": 0.0177,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.0354609929078014,
|
| 2016 |
+
"grad_norm": 0.7367897629737854,
|
| 2017 |
+
"learning_rate": 2.5737107174878772e-05,
|
| 2018 |
+
"loss": 0.0281,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.0425531914893615,
|
| 2023 |
+
"grad_norm": 0.29826951026916504,
|
| 2024 |
+
"learning_rate": 2.5396025785776545e-05,
|
| 2025 |
+
"loss": 0.0245,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.049645390070922,
|
| 2030 |
+
"grad_norm": 0.3840036988258362,
|
| 2031 |
+
"learning_rate": 2.505644851821633e-05,
|
| 2032 |
+
"loss": 0.0235,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.0567375886524824,
|
| 2037 |
+
"grad_norm": 0.43504858016967773,
|
| 2038 |
+
"learning_rate": 2.4718396131668876e-05,
|
| 2039 |
+
"loss": 0.0227,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.0638297872340425,
|
| 2044 |
+
"grad_norm": 0.5167884826660156,
|
| 2045 |
+
"learning_rate": 2.4381889292383997e-05,
|
| 2046 |
+
"loss": 0.0219,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.0709219858156027,
|
| 2051 |
+
"grad_norm": 0.38557201623916626,
|
| 2052 |
+
"learning_rate": 2.4046948572127076e-05,
|
| 2053 |
+
"loss": 0.0204,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.078014184397163,
|
| 2058 |
+
"grad_norm": 0.4263916015625,
|
| 2059 |
+
"learning_rate": 2.371359444692155e-05,
|
| 2060 |
+
"loss": 0.0239,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.0851063829787235,
|
| 2065 |
+
"grad_norm": 0.48733383417129517,
|
| 2066 |
+
"learning_rate": 2.3381847295797138e-05,
|
| 2067 |
+
"loss": 0.0245,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.0921985815602837,
|
| 2072 |
+
"grad_norm": 0.26661062240600586,
|
| 2073 |
+
"learning_rate": 2.3051727399543932e-05,
|
| 2074 |
+
"loss": 0.0241,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.099290780141844,
|
| 2079 |
+
"grad_norm": 0.48153558373451233,
|
| 2080 |
+
"learning_rate": 2.272325493947257e-05,
|
| 2081 |
+
"loss": 0.0217,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.106382978723404,
|
| 2086 |
+
"grad_norm": 0.3827018439769745,
|
| 2087 |
+
"learning_rate": 2.239644999618058e-05,
|
| 2088 |
+
"loss": 0.0214,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.1134751773049647,
|
| 2093 |
+
"grad_norm": 0.2948437035083771,
|
| 2094 |
+
"learning_rate": 2.2071332548324685e-05,
|
| 2095 |
+
"loss": 0.0194,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.120567375886525,
|
| 2100 |
+
"grad_norm": 0.32214516401290894,
|
| 2101 |
+
"learning_rate": 2.1747922471399516e-05,
|
| 2102 |
+
"loss": 0.0226,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.127659574468085,
|
| 2107 |
+
"grad_norm": 0.6189494729042053,
|
| 2108 |
+
"learning_rate": 2.1426239536522495e-05,
|
| 2109 |
+
"loss": 0.0239,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.1347517730496453,
|
| 2114 |
+
"grad_norm": 0.4420052468776703,
|
| 2115 |
+
"learning_rate": 2.1106303409225225e-05,
|
| 2116 |
+
"loss": 0.0231,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.141843971631206,
|
| 2121 |
+
"grad_norm": 0.5005484819412231,
|
| 2122 |
+
"learning_rate": 2.0788133648251206e-05,
|
| 2123 |
+
"loss": 0.0196,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.148936170212766,
|
| 2128 |
+
"grad_norm": 0.3406815528869629,
|
| 2129 |
+
"learning_rate": 2.0471749704360215e-05,
|
| 2130 |
+
"loss": 0.0236,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.1560283687943262,
|
| 2135 |
+
"grad_norm": 0.5321279168128967,
|
| 2136 |
+
"learning_rate": 2.0157170919139136e-05,
|
| 2137 |
+
"loss": 0.0209,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.1631205673758864,
|
| 2142 |
+
"grad_norm": 0.47004616260528564,
|
| 2143 |
+
"learning_rate": 1.984441652381962e-05,
|
| 2144 |
+
"loss": 0.0263,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.1702127659574466,
|
| 2149 |
+
"grad_norm": 0.361755907535553,
|
| 2150 |
+
"learning_rate": 1.9533505638102383e-05,
|
| 2151 |
+
"loss": 0.0189,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.1773049645390072,
|
| 2156 |
+
"grad_norm": 0.2968461513519287,
|
| 2157 |
+
"learning_rate": 1.9224457268988365e-05,
|
| 2158 |
+
"loss": 0.0196,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.1843971631205674,
|
| 2163 |
+
"grad_norm": 0.4293064475059509,
|
| 2164 |
+
"learning_rate": 1.891729030961675e-05,
|
| 2165 |
+
"loss": 0.0207,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.1914893617021276,
|
| 2170 |
+
"grad_norm": 0.3457898795604706,
|
| 2171 |
+
"learning_rate": 1.8612023538109996e-05,
|
| 2172 |
+
"loss": 0.0203,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.198581560283688,
|
| 2177 |
+
"grad_norm": 0.3276647627353668,
|
| 2178 |
+
"learning_rate": 1.830867561642584e-05,
|
| 2179 |
+
"loss": 0.0187,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.2056737588652484,
|
| 2184 |
+
"grad_norm": 0.33857762813568115,
|
| 2185 |
+
"learning_rate": 1.800726508921647e-05,
|
| 2186 |
+
"loss": 0.0243,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.2127659574468086,
|
| 2191 |
+
"grad_norm": 0.293497234582901,
|
| 2192 |
+
"learning_rate": 1.7707810382694745e-05,
|
| 2193 |
+
"loss": 0.024,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.219858156028369,
|
| 2198 |
+
"grad_norm": 0.392678439617157,
|
| 2199 |
+
"learning_rate": 1.741032980350786e-05,
|
| 2200 |
+
"loss": 0.0189,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.226950354609929,
|
| 2205 |
+
"grad_norm": 0.3549427390098572,
|
| 2206 |
+
"learning_rate": 1.7114841537618083e-05,
|
| 2207 |
+
"loss": 0.018,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.2340425531914896,
|
| 2212 |
+
"grad_norm": 0.554305911064148,
|
| 2213 |
+
"learning_rate": 1.682136364919112e-05,
|
| 2214 |
+
"loss": 0.0224,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.2411347517730498,
|
| 2219 |
+
"grad_norm": 0.5532910823822021,
|
| 2220 |
+
"learning_rate": 1.652991407949167e-05,
|
| 2221 |
+
"loss": 0.0284,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.24822695035461,
|
| 2226 |
+
"grad_norm": 0.4016483426094055,
|
| 2227 |
+
"learning_rate": 1.624051064578664e-05,
|
| 2228 |
+
"loss": 0.0202,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.25531914893617,
|
| 2233 |
+
"grad_norm": 0.33500346541404724,
|
| 2234 |
+
"learning_rate": 1.5953171040255965e-05,
|
| 2235 |
+
"loss": 0.021,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.2624113475177303,
|
| 2240 |
+
"grad_norm": 0.239231139421463,
|
| 2241 |
+
"learning_rate": 1.5667912828911025e-05,
|
| 2242 |
+
"loss": 0.0196,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.269503546099291,
|
| 2247 |
+
"grad_norm": 0.2729613780975342,
|
| 2248 |
+
"learning_rate": 1.5384753450520738e-05,
|
| 2249 |
+
"loss": 0.0226,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.276595744680851,
|
| 2254 |
+
"grad_norm": 0.26264944672584534,
|
| 2255 |
+
"learning_rate": 1.5103710215545446e-05,
|
| 2256 |
+
"loss": 0.0206,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.2836879432624113,
|
| 2261 |
+
"grad_norm": 0.37236642837524414,
|
| 2262 |
+
"learning_rate": 1.4824800305078795e-05,
|
| 2263 |
+
"loss": 0.0202,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.2907801418439715,
|
| 2268 |
+
"grad_norm": 0.6244298219680786,
|
| 2269 |
+
"learning_rate": 1.4548040769797256e-05,
|
| 2270 |
+
"loss": 0.0201,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.297872340425532,
|
| 2275 |
+
"grad_norm": 0.43154317140579224,
|
| 2276 |
+
"learning_rate": 1.4273448528917854e-05,
|
| 2277 |
+
"loss": 0.0192,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.3049645390070923,
|
| 2282 |
+
"grad_norm": 0.3106103241443634,
|
| 2283 |
+
"learning_rate": 1.4001040369163753e-05,
|
| 2284 |
+
"loss": 0.0231,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.3120567375886525,
|
| 2289 |
+
"grad_norm": 0.27755704522132874,
|
| 2290 |
+
"learning_rate": 1.3730832943738141e-05,
|
| 2291 |
+
"loss": 0.0209,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.3191489361702127,
|
| 2296 |
+
"grad_norm": 0.7556533217430115,
|
| 2297 |
+
"learning_rate": 1.3462842771306083e-05,
|
| 2298 |
+
"loss": 0.0188,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.326241134751773,
|
| 2303 |
+
"grad_norm": 0.5042951703071594,
|
| 2304 |
+
"learning_rate": 1.3197086234984706e-05,
|
| 2305 |
+
"loss": 0.0205,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.3333333333333335,
|
| 2310 |
+
"grad_norm": 0.3070732355117798,
|
| 2311 |
+
"learning_rate": 1.293357958134166e-05,
|
| 2312 |
+
"loss": 0.0203,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.3404255319148937,
|
| 2317 |
+
"grad_norm": 0.3759100139141083,
|
| 2318 |
+
"learning_rate": 1.2672338919401866e-05,
|
| 2319 |
+
"loss": 0.0169,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.347517730496454,
|
| 2324 |
+
"grad_norm": 0.2854216396808624,
|
| 2325 |
+
"learning_rate": 1.241338021966278e-05,
|
| 2326 |
+
"loss": 0.019,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.354609929078014,
|
| 2331 |
+
"grad_norm": 0.3408313989639282,
|
| 2332 |
+
"learning_rate": 1.2156719313118026e-05,
|
| 2333 |
+
"loss": 0.0173,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.3617021276595747,
|
| 2338 |
+
"grad_norm": 0.55415940284729,
|
| 2339 |
+
"learning_rate": 1.190237189028957e-05,
|
| 2340 |
+
"loss": 0.0211,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.368794326241135,
|
| 2345 |
+
"grad_norm": 0.39499035477638245,
|
| 2346 |
+
"learning_rate": 1.165035350026859e-05,
|
| 2347 |
+
"loss": 0.0182,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.375886524822695,
|
| 2352 |
+
"grad_norm": 0.36144962906837463,
|
| 2353 |
+
"learning_rate": 1.1400679549764825e-05,
|
| 2354 |
+
"loss": 0.0227,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.382978723404255,
|
| 2359 |
+
"grad_norm": 0.34371185302734375,
|
| 2360 |
+
"learning_rate": 1.1153365302164765e-05,
|
| 2361 |
+
"loss": 0.0176,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.3900709219858154,
|
| 2366 |
+
"grad_norm": 0.2524702250957489,
|
| 2367 |
+
"learning_rate": 1.090842587659851e-05,
|
| 2368 |
+
"loss": 0.0191,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.397163120567376,
|
| 2373 |
+
"grad_norm": 0.3390008211135864,
|
| 2374 |
+
"learning_rate": 1.0665876247015544e-05,
|
| 2375 |
+
"loss": 0.0158,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.404255319148936,
|
| 2380 |
+
"grad_norm": 0.2771461308002472,
|
| 2381 |
+
"learning_rate": 1.0425731241269255e-05,
|
| 2382 |
+
"loss": 0.0171,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.4113475177304964,
|
| 2387 |
+
"grad_norm": 0.4448011815547943,
|
| 2388 |
+
"learning_rate": 1.0188005540210543e-05,
|
| 2389 |
+
"loss": 0.0212,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.4184397163120566,
|
| 2394 |
+
"grad_norm": 0.49088287353515625,
|
| 2395 |
+
"learning_rate": 9.952713676790225e-06,
|
| 2396 |
+
"loss": 0.0202,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.425531914893617,
|
| 2401 |
+
"grad_norm": 0.2753278911113739,
|
| 2402 |
+
"learning_rate": 9.719870035170698e-06,
|
| 2403 |
+
"loss": 0.0174,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.4326241134751774,
|
| 2408 |
+
"grad_norm": 0.3511912524700165,
|
| 2409 |
+
"learning_rate": 9.48948884984654e-06,
|
| 2410 |
+
"loss": 0.0203,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.4397163120567376,
|
| 2415 |
+
"grad_norm": 0.38820111751556396,
|
| 2416 |
+
"learning_rate": 9.26158420477431e-06,
|
| 2417 |
+
"loss": 0.0214,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.4468085106382977,
|
| 2422 |
+
"grad_norm": 0.27879631519317627,
|
| 2423 |
+
"learning_rate": 9.03617003251151e-06,
|
| 2424 |
+
"loss": 0.016,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.453900709219858,
|
| 2429 |
+
"grad_norm": 0.2823965549468994,
|
| 2430 |
+
"learning_rate": 8.813260113364912e-06,
|
| 2431 |
+
"loss": 0.0204,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.4609929078014185,
|
| 2436 |
+
"grad_norm": 0.35228607058525085,
|
| 2437 |
+
"learning_rate": 8.5928680745481e-06,
|
| 2438 |
+
"loss": 0.02,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.4680851063829787,
|
| 2443 |
+
"grad_norm": 0.40652936697006226,
|
| 2444 |
+
"learning_rate": 8.37500738934836e-06,
|
| 2445 |
+
"loss": 0.0192,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.475177304964539,
|
| 2450 |
+
"grad_norm": 0.3318992257118225,
|
| 2451 |
+
"learning_rate": 8.159691376303058e-06,
|
| 2452 |
+
"loss": 0.0204,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.482269503546099,
|
| 2457 |
+
"grad_norm": 0.37183699011802673,
|
| 2458 |
+
"learning_rate": 7.94693319838541e-06,
|
| 2459 |
+
"loss": 0.0163,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.4893617021276597,
|
| 2464 |
+
"grad_norm": 0.7285052537918091,
|
| 2465 |
+
"learning_rate": 7.736745862199785e-06,
|
| 2466 |
+
"loss": 0.0267,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.49645390070922,
|
| 2471 |
+
"grad_norm": 0.6088675856590271,
|
| 2472 |
+
"learning_rate": 7.529142217186597e-06,
|
| 2473 |
+
"loss": 0.0209,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.50354609929078,
|
| 2478 |
+
"grad_norm": 0.5769776701927185,
|
| 2479 |
+
"learning_rate": 7.324134954836709e-06,
|
| 2480 |
+
"loss": 0.0193,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.5106382978723403,
|
| 2485 |
+
"grad_norm": 0.2930261492729187,
|
| 2486 |
+
"learning_rate": 7.121736607915658e-06,
|
| 2487 |
+
"loss": 0.016,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.5177304964539005,
|
| 2492 |
+
"grad_norm": 0.3488508462905884,
|
| 2493 |
+
"learning_rate": 6.921959549697404e-06,
|
| 2494 |
+
"loss": 0.0172,
|
| 2495 |
+
"step": 3550
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.524822695035461,
|
| 2499 |
+
"grad_norm": 0.23564404249191284,
|
| 2500 |
+
"learning_rate": 6.724815993207955e-06,
|
| 2501 |
+
"loss": 0.0208,
|
| 2502 |
+
"step": 3560
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 2.5319148936170213,
|
| 2506 |
+
"grad_norm": 0.3507620692253113,
|
| 2507 |
+
"learning_rate": 6.53031799047873e-06,
|
| 2508 |
+
"loss": 0.0172,
|
| 2509 |
+
"step": 3570
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 2.5390070921985815,
|
| 2513 |
+
"grad_norm": 0.463963121175766,
|
| 2514 |
+
"learning_rate": 6.338477431809764e-06,
|
| 2515 |
+
"loss": 0.0178,
|
| 2516 |
+
"step": 3580
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 2.546099290780142,
|
| 2520 |
+
"grad_norm": 0.4759658873081207,
|
| 2521 |
+
"learning_rate": 6.149306045042858e-06,
|
| 2522 |
+
"loss": 0.0162,
|
| 2523 |
+
"step": 3590
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 2.5531914893617023,
|
| 2527 |
+
"grad_norm": 0.3458139896392822,
|
| 2528 |
+
"learning_rate": 5.962815394844568e-06,
|
| 2529 |
+
"loss": 0.0192,
|
| 2530 |
+
"step": 3600
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 2.5602836879432624,
|
| 2534 |
+
"grad_norm": 0.3261510729789734,
|
| 2535 |
+
"learning_rate": 5.7790168819992666e-06,
|
| 2536 |
+
"loss": 0.0176,
|
| 2537 |
+
"step": 3610
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 2.5673758865248226,
|
| 2541 |
+
"grad_norm": 0.24720831215381622,
|
| 2542 |
+
"learning_rate": 5.597921742712115e-06,
|
| 2543 |
+
"loss": 0.0187,
|
| 2544 |
+
"step": 3620
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 2.574468085106383,
|
| 2548 |
+
"grad_norm": 0.24207931756973267,
|
| 2549 |
+
"learning_rate": 5.419541047922216e-06,
|
| 2550 |
+
"loss": 0.0159,
|
| 2551 |
+
"step": 3630
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 2.581560283687943,
|
| 2555 |
+
"grad_norm": 0.2809526026248932,
|
| 2556 |
+
"learning_rate": 5.243885702625795e-06,
|
| 2557 |
+
"loss": 0.017,
|
| 2558 |
+
"step": 3640
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 2.5886524822695036,
|
| 2562 |
+
"grad_norm": 0.3030683696269989,
|
| 2563 |
+
"learning_rate": 5.070966445209541e-06,
|
| 2564 |
+
"loss": 0.0211,
|
| 2565 |
+
"step": 3650
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 2.595744680851064,
|
| 2569 |
+
"grad_norm": 0.4494501054286957,
|
| 2570 |
+
"learning_rate": 4.900793846794077e-06,
|
| 2571 |
+
"loss": 0.0209,
|
| 2572 |
+
"step": 3660
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 2.602836879432624,
|
| 2576 |
+
"grad_norm": 0.2733117640018463,
|
| 2577 |
+
"learning_rate": 4.733378310587827e-06,
|
| 2578 |
+
"loss": 0.017,
|
| 2579 |
+
"step": 3670
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 2.6099290780141846,
|
| 2583 |
+
"grad_norm": 0.21026711165905,
|
| 2584 |
+
"learning_rate": 4.568730071250926e-06,
|
| 2585 |
+
"loss": 0.0183,
|
| 2586 |
+
"step": 3680
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 2.617021276595745,
|
| 2590 |
+
"grad_norm": 0.24560986459255219,
|
| 2591 |
+
"learning_rate": 4.406859194269619e-06,
|
| 2592 |
+
"loss": 0.0191,
|
| 2593 |
+
"step": 3690
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 2.624113475177305,
|
| 2597 |
+
"grad_norm": 0.2999397814273834,
|
| 2598 |
+
"learning_rate": 4.247775575340862e-06,
|
| 2599 |
+
"loss": 0.0186,
|
| 2600 |
+
"step": 3700
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 2.631205673758865,
|
| 2604 |
+
"grad_norm": 0.322882741689682,
|
| 2605 |
+
"learning_rate": 4.091488939767424e-06,
|
| 2606 |
+
"loss": 0.0223,
|
| 2607 |
+
"step": 3710
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 2.6382978723404253,
|
| 2611 |
+
"grad_norm": 0.33875447511672974,
|
| 2612 |
+
"learning_rate": 3.938008841863289e-06,
|
| 2613 |
+
"loss": 0.0218,
|
| 2614 |
+
"step": 3720
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 2.645390070921986,
|
| 2618 |
+
"grad_norm": 0.29696741700172424,
|
| 2619 |
+
"learning_rate": 3.787344664369641e-06,
|
| 2620 |
+
"loss": 0.018,
|
| 2621 |
+
"step": 3730
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 2.652482269503546,
|
| 2625 |
+
"grad_norm": 0.3487037718296051,
|
| 2626 |
+
"learning_rate": 3.6395056178811726e-06,
|
| 2627 |
+
"loss": 0.0231,
|
| 2628 |
+
"step": 3740
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 2.6595744680851063,
|
| 2632 |
+
"grad_norm": 0.4687223732471466,
|
| 2633 |
+
"learning_rate": 3.4945007402830965e-06,
|
| 2634 |
+
"loss": 0.0207,
|
| 2635 |
+
"step": 3750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 2.6666666666666665,
|
| 2639 |
+
"grad_norm": 0.38484394550323486,
|
| 2640 |
+
"learning_rate": 3.352338896198598e-06,
|
| 2641 |
+
"loss": 0.0154,
|
| 2642 |
+
"step": 3760
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 2.673758865248227,
|
| 2646 |
+
"grad_norm": 0.28347933292388916,
|
| 2647 |
+
"learning_rate": 3.2130287764469024e-06,
|
| 2648 |
+
"loss": 0.0192,
|
| 2649 |
+
"step": 3770
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 2.6808510638297873,
|
| 2653 |
+
"grad_norm": 0.2538928687572479,
|
| 2654 |
+
"learning_rate": 3.076578897511978e-06,
|
| 2655 |
+
"loss": 0.0175,
|
| 2656 |
+
"step": 3780
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 2.6879432624113475,
|
| 2660 |
+
"grad_norm": 0.27914267778396606,
|
| 2661 |
+
"learning_rate": 2.942997601021924e-06,
|
| 2662 |
+
"loss": 0.02,
|
| 2663 |
+
"step": 3790
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 2.6950354609929077,
|
| 2667 |
+
"grad_norm": 0.4712008237838745,
|
| 2668 |
+
"learning_rate": 2.812293053238968e-06,
|
| 2669 |
+
"loss": 0.017,
|
| 2670 |
+
"step": 3800
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 2.702127659574468,
|
| 2674 |
+
"grad_norm": 0.29678672552108765,
|
| 2675 |
+
"learning_rate": 2.684473244560298e-06,
|
| 2676 |
+
"loss": 0.0205,
|
| 2677 |
+
"step": 3810
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 2.7092198581560285,
|
| 2681 |
+
"grad_norm": 0.4024706184864044,
|
| 2682 |
+
"learning_rate": 2.5595459890295103e-06,
|
| 2683 |
+
"loss": 0.0201,
|
| 2684 |
+
"step": 3820
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 2.7163120567375887,
|
| 2688 |
+
"grad_norm": 0.17329274117946625,
|
| 2689 |
+
"learning_rate": 2.4375189238589944e-06,
|
| 2690 |
+
"loss": 0.0182,
|
| 2691 |
+
"step": 3830
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 2.723404255319149,
|
| 2695 |
+
"grad_norm": 0.32226958870887756,
|
| 2696 |
+
"learning_rate": 2.3183995089629706e-06,
|
| 2697 |
+
"loss": 0.0167,
|
| 2698 |
+
"step": 3840
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 2.7304964539007095,
|
| 2702 |
+
"grad_norm": 0.23052819073200226,
|
| 2703 |
+
"learning_rate": 2.2021950265015077e-06,
|
| 2704 |
+
"loss": 0.0191,
|
| 2705 |
+
"step": 3850
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 2.7375886524822697,
|
| 2709 |
+
"grad_norm": 0.1971309781074524,
|
| 2710 |
+
"learning_rate": 2.0889125804352592e-06,
|
| 2711 |
+
"loss": 0.0175,
|
| 2712 |
+
"step": 3860
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 2.74468085106383,
|
| 2716 |
+
"grad_norm": 0.3845289349555969,
|
| 2717 |
+
"learning_rate": 1.9785590960912537e-06,
|
| 2718 |
+
"loss": 0.0183,
|
| 2719 |
+
"step": 3870
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 2.75177304964539,
|
| 2723 |
+
"grad_norm": 0.37887126207351685,
|
| 2724 |
+
"learning_rate": 1.8711413197394945e-06,
|
| 2725 |
+
"loss": 0.0169,
|
| 2726 |
+
"step": 3880
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 2.7588652482269502,
|
| 2730 |
+
"grad_norm": 0.28578147292137146,
|
| 2731 |
+
"learning_rate": 1.7666658181805297e-06,
|
| 2732 |
+
"loss": 0.0143,
|
| 2733 |
+
"step": 3890
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 2.7659574468085104,
|
| 2737 |
+
"grad_norm": 0.2607921063899994,
|
| 2738 |
+
"learning_rate": 1.6651389783440109e-06,
|
| 2739 |
+
"loss": 0.0158,
|
| 2740 |
+
"step": 3900
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 2.773049645390071,
|
| 2744 |
+
"grad_norm": 0.3524990379810333,
|
| 2745 |
+
"learning_rate": 1.5665670068982286e-06,
|
| 2746 |
+
"loss": 0.0184,
|
| 2747 |
+
"step": 3910
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 2.780141843971631,
|
| 2751 |
+
"grad_norm": 0.19082817435264587,
|
| 2752 |
+
"learning_rate": 1.4709559298707265e-06,
|
| 2753 |
+
"loss": 0.0165,
|
| 2754 |
+
"step": 3920
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 2.7872340425531914,
|
| 2758 |
+
"grad_norm": 0.6385450959205627,
|
| 2759 |
+
"learning_rate": 1.378311592279835e-06,
|
| 2760 |
+
"loss": 0.0154,
|
| 2761 |
+
"step": 3930
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 2.794326241134752,
|
| 2765 |
+
"grad_norm": 0.336570680141449,
|
| 2766 |
+
"learning_rate": 1.2886396577773963e-06,
|
| 2767 |
+
"loss": 0.0166,
|
| 2768 |
+
"step": 3940
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 2.801418439716312,
|
| 2772 |
+
"grad_norm": 0.2245451658964157,
|
| 2773 |
+
"learning_rate": 1.2019456083025182e-06,
|
| 2774 |
+
"loss": 0.0154,
|
| 2775 |
+
"step": 3950
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 2.8085106382978724,
|
| 2779 |
+
"grad_norm": 0.4137704074382782,
|
| 2780 |
+
"learning_rate": 1.1182347437464436e-06,
|
| 2781 |
+
"loss": 0.0127,
|
| 2782 |
+
"step": 3960
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 2.8156028368794326,
|
| 2786 |
+
"grad_norm": 0.6312252283096313,
|
| 2787 |
+
"learning_rate": 1.037512181628536e-06,
|
| 2788 |
+
"loss": 0.0173,
|
| 2789 |
+
"step": 3970
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 2.8226950354609928,
|
| 2793 |
+
"grad_norm": 0.2965438663959503,
|
| 2794 |
+
"learning_rate": 9.597828567834589e-07,
|
| 2795 |
+
"loss": 0.0191,
|
| 2796 |
+
"step": 3980
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 2.829787234042553,
|
| 2800 |
+
"grad_norm": 0.22976373136043549,
|
| 2801 |
+
"learning_rate": 8.850515210594601e-07,
|
| 2802 |
+
"loss": 0.0132,
|
| 2803 |
+
"step": 3990
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 2.8368794326241136,
|
| 2807 |
+
"grad_norm": 0.2823370695114136,
|
| 2808 |
+
"learning_rate": 8.133227430279055e-07,
|
| 2809 |
+
"loss": 0.0158,
|
| 2810 |
+
"step": 4000
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 2.8439716312056738,
|
| 2814 |
+
"grad_norm": 0.3871881067752838,
|
| 2815 |
+
"learning_rate": 7.446009077039629e-07,
|
| 2816 |
+
"loss": 0.0161,
|
| 2817 |
+
"step": 4010
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 2.851063829787234,
|
| 2821 |
+
"grad_norm": 0.39857760071754456,
|
| 2822 |
+
"learning_rate": 6.788902162785549e-07,
|
| 2823 |
+
"loss": 0.016,
|
| 2824 |
+
"step": 4020
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 2.8581560283687946,
|
| 2828 |
+
"grad_norm": 0.32643163204193115,
|
| 2829 |
+
"learning_rate": 6.161946858614931e-07,
|
| 2830 |
+
"loss": 0.0177,
|
| 2831 |
+
"step": 4030
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 2.8652482269503547,
|
| 2835 |
+
"grad_norm": 0.397289901971817,
|
| 2836 |
+
"learning_rate": 5.565181492359406e-07,
|
| 2837 |
+
"loss": 0.019,
|
| 2838 |
+
"step": 4040
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 2.872340425531915,
|
| 2842 |
+
"grad_norm": 0.20939776301383972,
|
| 2843 |
+
"learning_rate": 4.998642546240606e-07,
|
| 2844 |
+
"loss": 0.0179,
|
| 2845 |
+
"step": 4050
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 2.879432624113475,
|
| 2849 |
+
"grad_norm": 0.20747888088226318,
|
| 2850 |
+
"learning_rate": 4.462364654640283e-07,
|
| 2851 |
+
"loss": 0.0169,
|
| 2852 |
+
"step": 4060
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 2.8865248226950353,
|
| 2856 |
+
"grad_norm": 0.5459475517272949,
|
| 2857 |
+
"learning_rate": 3.9563806019826677e-07,
|
| 2858 |
+
"loss": 0.0188,
|
| 2859 |
+
"step": 4070
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 2.8936170212765955,
|
| 2863 |
+
"grad_norm": 0.3425324261188507,
|
| 2864 |
+
"learning_rate": 3.4807213207304624e-07,
|
| 2865 |
+
"loss": 0.0173,
|
| 2866 |
+
"step": 4080
|
| 2867 |
+
},
|
| 2868 |
+
{
|
| 2869 |
+
"epoch": 2.900709219858156,
|
| 2870 |
+
"grad_norm": 0.4077880382537842,
|
| 2871 |
+
"learning_rate": 3.035415889493631e-07,
|
| 2872 |
+
"loss": 0.0192,
|
| 2873 |
+
"step": 4090
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 2.9078014184397163,
|
| 2877 |
+
"grad_norm": 0.5677205920219421,
|
| 2878 |
+
"learning_rate": 2.62049153125199e-07,
|
| 2879 |
+
"loss": 0.0199,
|
| 2880 |
+
"step": 4100
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 2.9148936170212765,
|
| 2884 |
+
"grad_norm": 0.24135500192642212,
|
| 2885 |
+
"learning_rate": 2.2359736116907005e-07,
|
| 2886 |
+
"loss": 0.0191,
|
| 2887 |
+
"step": 4110
|
| 2888 |
+
},
|
| 2889 |
+
{
|
| 2890 |
+
"epoch": 2.921985815602837,
|
| 2891 |
+
"grad_norm": 0.43725013732910156,
|
| 2892 |
+
"learning_rate": 1.8818856376498494e-07,
|
| 2893 |
+
"loss": 0.0184,
|
| 2894 |
+
"step": 4120
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 2.9290780141843973,
|
| 2898 |
+
"grad_norm": 0.3367736339569092,
|
| 2899 |
+
"learning_rate": 1.5582492556872608e-07,
|
| 2900 |
+
"loss": 0.0172,
|
| 2901 |
+
"step": 4130
|
| 2902 |
+
},
|
| 2903 |
+
{
|
| 2904 |
+
"epoch": 2.9361702127659575,
|
| 2905 |
+
"grad_norm": 0.43944522738456726,
|
| 2906 |
+
"learning_rate": 1.2650842507550554e-07,
|
| 2907 |
+
"loss": 0.0183,
|
| 2908 |
+
"step": 4140
|
| 2909 |
+
},
|
| 2910 |
+
{
|
| 2911 |
+
"epoch": 2.9432624113475176,
|
| 2912 |
+
"grad_norm": 0.29025810956954956,
|
| 2913 |
+
"learning_rate": 1.0024085449903964e-07,
|
| 2914 |
+
"loss": 0.0138,
|
| 2915 |
+
"step": 4150
|
| 2916 |
+
},
|
| 2917 |
+
{
|
| 2918 |
+
"epoch": 2.950354609929078,
|
| 2919 |
+
"grad_norm": 0.6834773421287537,
|
| 2920 |
+
"learning_rate": 7.702381966196437e-08,
|
| 2921 |
+
"loss": 0.0161,
|
| 2922 |
+
"step": 4160
|
| 2923 |
+
},
|
| 2924 |
+
{
|
| 2925 |
+
"epoch": 2.9574468085106385,
|
| 2926 |
+
"grad_norm": 0.34878963232040405,
|
| 2927 |
+
"learning_rate": 5.685873989767498e-08,
|
| 2928 |
+
"loss": 0.0162,
|
| 2929 |
+
"step": 4170
|
| 2930 |
+
},
|
| 2931 |
+
{
|
| 2932 |
+
"epoch": 2.9645390070921986,
|
| 2933 |
+
"grad_norm": 0.3399132192134857,
|
| 2934 |
+
"learning_rate": 3.974684796355099e-08,
|
| 2935 |
+
"loss": 0.0224,
|
| 2936 |
+
"step": 4180
|
| 2937 |
+
},
|
| 2938 |
+
{
|
| 2939 |
+
"epoch": 2.971631205673759,
|
| 2940 |
+
"grad_norm": 0.48755013942718506,
|
| 2941 |
+
"learning_rate": 2.568918996560532e-08,
|
| 2942 |
+
"loss": 0.0161,
|
| 2943 |
+
"step": 4190
|
| 2944 |
+
},
|
| 2945 |
+
{
|
| 2946 |
+
"epoch": 2.978723404255319,
|
| 2947 |
+
"grad_norm": 0.3584892153739929,
|
| 2948 |
+
"learning_rate": 1.468662529452436e-08,
|
| 2949 |
+
"loss": 0.0183,
|
| 2950 |
+
"step": 4200
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 2.9858156028368796,
|
| 2954 |
+
"grad_norm": 0.3036380112171173,
|
| 2955 |
+
"learning_rate": 6.7398265731211105e-09,
|
| 2956 |
+
"loss": 0.0147,
|
| 2957 |
+
"step": 4210
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 2.99290780141844,
|
| 2961 |
+
"grad_norm": 0.3986826539039612,
|
| 2962 |
+
"learning_rate": 1.8492796152402847e-09,
|
| 2963 |
+
"loss": 0.0166,
|
| 2964 |
+
"step": 4220
|
| 2965 |
+
},
|
| 2966 |
+
{
|
| 2967 |
+
"epoch": 3.0,
|
| 2968 |
+
"grad_norm": 0.21891933679580688,
|
| 2969 |
+
"learning_rate": 1.5283396037624188e-11,
|
| 2970 |
+
"loss": 0.0165,
|
| 2971 |
+
"step": 4230
|
| 2972 |
+
},
|
| 2973 |
+
{
|
| 2974 |
+
"epoch": 3.0,
|
| 2975 |
+
"step": 4230,
|
| 2976 |
+
"total_flos": 0.0,
|
| 2977 |
+
"train_loss": 0.03590505794219091,
|
| 2978 |
+
"train_runtime": 4518.8322,
|
| 2979 |
+
"train_samples_per_second": 45.842,
|
| 2980 |
+
"train_steps_per_second": 0.936
|
| 2981 |
+
}
|
| 2982 |
+
],
|
| 2983 |
+
"logging_steps": 10,
|
| 2984 |
+
"max_steps": 4230,
|
| 2985 |
+
"num_input_tokens_seen": 0,
|
| 2986 |
+
"num_train_epochs": 3,
|
| 2987 |
+
"save_steps": 5000,
|
| 2988 |
+
"stateful_callbacks": {
|
| 2989 |
+
"TrainerControl": {
|
| 2990 |
+
"args": {
|
| 2991 |
+
"should_epoch_stop": false,
|
| 2992 |
+
"should_evaluate": false,
|
| 2993 |
+
"should_log": false,
|
| 2994 |
+
"should_save": true,
|
| 2995 |
+
"should_training_stop": true
|
| 2996 |
+
},
|
| 2997 |
+
"attributes": {}
|
| 2998 |
+
}
|
| 2999 |
+
},
|
| 3000 |
+
"total_flos": 0.0,
|
| 3001 |
+
"train_batch_size": 49,
|
| 3002 |
+
"trial_name": null,
|
| 3003 |
+
"trial_params": null
|
| 3004 |
+
}
|