Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +1442 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1,
|
| 13 |
+
"grad_norm": 8.593878746032715,
|
| 14 |
+
"learning_rate": 1e-05,
|
| 15 |
+
"loss": 1.4852,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2,
|
| 20 |
+
"grad_norm": 7.231457233428955,
|
| 21 |
+
"learning_rate": 2e-05,
|
| 22 |
+
"loss": 1.4207,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3,
|
| 27 |
+
"grad_norm": 3.319319248199463,
|
| 28 |
+
"learning_rate": 3e-05,
|
| 29 |
+
"loss": 1.0614,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4,
|
| 34 |
+
"grad_norm": 3.60310435295105,
|
| 35 |
+
"learning_rate": 4e-05,
|
| 36 |
+
"loss": 0.7256,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5,
|
| 41 |
+
"grad_norm": 15.761638641357422,
|
| 42 |
+
"learning_rate": 5e-05,
|
| 43 |
+
"loss": 0.5587,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6,
|
| 48 |
+
"grad_norm": 24.317516326904297,
|
| 49 |
+
"learning_rate": 6e-05,
|
| 50 |
+
"loss": 0.5146,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7,
|
| 55 |
+
"grad_norm": 4.110610008239746,
|
| 56 |
+
"learning_rate": 7e-05,
|
| 57 |
+
"loss": 0.4222,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8,
|
| 62 |
+
"grad_norm": 4.141885757446289,
|
| 63 |
+
"learning_rate": 8e-05,
|
| 64 |
+
"loss": 0.4355,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9,
|
| 69 |
+
"grad_norm": 3.7163727283477783,
|
| 70 |
+
"learning_rate": 9e-05,
|
| 71 |
+
"loss": 0.3939,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"grad_norm": 8.189441680908203,
|
| 77 |
+
"learning_rate": 0.0001,
|
| 78 |
+
"loss": 0.3712,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.1,
|
| 83 |
+
"grad_norm": 2.9945340156555176,
|
| 84 |
+
"learning_rate": 9.999316524962345e-05,
|
| 85 |
+
"loss": 0.3548,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.2,
|
| 90 |
+
"grad_norm": 1.997642159461975,
|
| 91 |
+
"learning_rate": 9.997266286704631e-05,
|
| 92 |
+
"loss": 0.2898,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.3,
|
| 97 |
+
"grad_norm": 1.9837431907653809,
|
| 98 |
+
"learning_rate": 9.993849845741524e-05,
|
| 99 |
+
"loss": 0.2528,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.4,
|
| 104 |
+
"grad_norm": 2.294177532196045,
|
| 105 |
+
"learning_rate": 9.989068136093873e-05,
|
| 106 |
+
"loss": 0.2187,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.5,
|
| 111 |
+
"grad_norm": 1.4083565473556519,
|
| 112 |
+
"learning_rate": 9.98292246503335e-05,
|
| 113 |
+
"loss": 0.1986,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.6,
|
| 118 |
+
"grad_norm": 7.1001811027526855,
|
| 119 |
+
"learning_rate": 9.975414512725057e-05,
|
| 120 |
+
"loss": 0.1738,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.7,
|
| 125 |
+
"grad_norm": 1.4144936800003052,
|
| 126 |
+
"learning_rate": 9.966546331768191e-05,
|
| 127 |
+
"loss": 0.1537,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.8,
|
| 132 |
+
"grad_norm": 1.5691213607788086,
|
| 133 |
+
"learning_rate": 9.956320346634876e-05,
|
| 134 |
+
"loss": 0.1519,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.9,
|
| 139 |
+
"grad_norm": 1.2897148132324219,
|
| 140 |
+
"learning_rate": 9.944739353007344e-05,
|
| 141 |
+
"loss": 0.1287,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.0,
|
| 146 |
+
"grad_norm": 1.6297178268432617,
|
| 147 |
+
"learning_rate": 9.931806517013612e-05,
|
| 148 |
+
"loss": 0.1169,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.1,
|
| 153 |
+
"grad_norm": 1.364025354385376,
|
| 154 |
+
"learning_rate": 9.917525374361912e-05,
|
| 155 |
+
"loss": 0.1246,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 2.2,
|
| 160 |
+
"grad_norm": 1.02783203125,
|
| 161 |
+
"learning_rate": 9.901899829374047e-05,
|
| 162 |
+
"loss": 0.1205,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 2.3,
|
| 167 |
+
"grad_norm": 1.1652231216430664,
|
| 168 |
+
"learning_rate": 9.884934153917997e-05,
|
| 169 |
+
"loss": 0.1123,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.4,
|
| 174 |
+
"grad_norm": 1.3376855850219727,
|
| 175 |
+
"learning_rate": 9.86663298624003e-05,
|
| 176 |
+
"loss": 0.1073,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.5,
|
| 181 |
+
"grad_norm": 0.9518831968307495,
|
| 182 |
+
"learning_rate": 9.847001329696653e-05,
|
| 183 |
+
"loss": 0.0983,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 2.6,
|
| 188 |
+
"grad_norm": 1.0183520317077637,
|
| 189 |
+
"learning_rate": 9.826044551386744e-05,
|
| 190 |
+
"loss": 0.0994,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 2.7,
|
| 195 |
+
"grad_norm": 1.088984727859497,
|
| 196 |
+
"learning_rate": 9.803768380684242e-05,
|
| 197 |
+
"loss": 0.1057,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.8,
|
| 202 |
+
"grad_norm": 0.7311228513717651,
|
| 203 |
+
"learning_rate": 9.780178907671789e-05,
|
| 204 |
+
"loss": 0.0979,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.9,
|
| 209 |
+
"grad_norm": 0.4584773778915405,
|
| 210 |
+
"learning_rate": 9.755282581475769e-05,
|
| 211 |
+
"loss": 0.1015,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.0,
|
| 216 |
+
"grad_norm": 1.5885437726974487,
|
| 217 |
+
"learning_rate": 9.729086208503174e-05,
|
| 218 |
+
"loss": 0.0798,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 3.1,
|
| 223 |
+
"grad_norm": 0.7659353017807007,
|
| 224 |
+
"learning_rate": 9.701596950580806e-05,
|
| 225 |
+
"loss": 0.0978,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 3.2,
|
| 230 |
+
"grad_norm": 0.472594290971756,
|
| 231 |
+
"learning_rate": 9.672822322997305e-05,
|
| 232 |
+
"loss": 0.0849,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.3,
|
| 237 |
+
"grad_norm": 0.9068973064422607,
|
| 238 |
+
"learning_rate": 9.642770192448536e-05,
|
| 239 |
+
"loss": 0.0917,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.4,
|
| 244 |
+
"grad_norm": 0.7755366563796997,
|
| 245 |
+
"learning_rate": 9.611448774886924e-05,
|
| 246 |
+
"loss": 0.0843,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 3.5,
|
| 251 |
+
"grad_norm": 1.4321980476379395,
|
| 252 |
+
"learning_rate": 9.578866633275288e-05,
|
| 253 |
+
"loss": 0.0813,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.6,
|
| 258 |
+
"grad_norm": 0.7324996590614319,
|
| 259 |
+
"learning_rate": 9.545032675245813e-05,
|
| 260 |
+
"loss": 0.0822,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 3.7,
|
| 265 |
+
"grad_norm": 0.5374558568000793,
|
| 266 |
+
"learning_rate": 9.509956150664796e-05,
|
| 267 |
+
"loss": 0.0877,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 3.8,
|
| 272 |
+
"grad_norm": 0.7615280151367188,
|
| 273 |
+
"learning_rate": 9.473646649103818e-05,
|
| 274 |
+
"loss": 0.0745,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 3.9,
|
| 279 |
+
"grad_norm": 1.008134365081787,
|
| 280 |
+
"learning_rate": 9.43611409721806e-05,
|
| 281 |
+
"loss": 0.0739,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 4.0,
|
| 286 |
+
"grad_norm": 1.376315951347351,
|
| 287 |
+
"learning_rate": 9.397368756032445e-05,
|
| 288 |
+
"loss": 0.0716,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 4.1,
|
| 293 |
+
"grad_norm": 1.1624388694763184,
|
| 294 |
+
"learning_rate": 9.357421218136386e-05,
|
| 295 |
+
"loss": 0.0787,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.2,
|
| 300 |
+
"grad_norm": 0.5839523077011108,
|
| 301 |
+
"learning_rate": 9.316282404787871e-05,
|
| 302 |
+
"loss": 0.0786,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.3,
|
| 307 |
+
"grad_norm": 0.9589359760284424,
|
| 308 |
+
"learning_rate": 9.273963562927695e-05,
|
| 309 |
+
"loss": 0.0735,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 4.4,
|
| 314 |
+
"grad_norm": 1.2537989616394043,
|
| 315 |
+
"learning_rate": 9.230476262104677e-05,
|
| 316 |
+
"loss": 0.0787,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 4.5,
|
| 321 |
+
"grad_norm": 0.9431202411651611,
|
| 322 |
+
"learning_rate": 9.185832391312644e-05,
|
| 323 |
+
"loss": 0.0772,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 4.6,
|
| 328 |
+
"grad_norm": 0.689441978931427,
|
| 329 |
+
"learning_rate": 9.140044155740101e-05,
|
| 330 |
+
"loss": 0.0775,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 4.7,
|
| 335 |
+
"grad_norm": 0.6225442290306091,
|
| 336 |
+
"learning_rate": 9.093124073433463e-05,
|
| 337 |
+
"loss": 0.0838,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 4.8,
|
| 342 |
+
"grad_norm": 0.8038715720176697,
|
| 343 |
+
"learning_rate": 9.045084971874738e-05,
|
| 344 |
+
"loss": 0.0708,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 4.9,
|
| 349 |
+
"grad_norm": 1.0997047424316406,
|
| 350 |
+
"learning_rate": 8.995939984474624e-05,
|
| 351 |
+
"loss": 0.0736,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 5.0,
|
| 356 |
+
"grad_norm": 1.7090861797332764,
|
| 357 |
+
"learning_rate": 8.945702546981969e-05,
|
| 358 |
+
"loss": 0.0664,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 5.1,
|
| 363 |
+
"grad_norm": 0.6094951629638672,
|
| 364 |
+
"learning_rate": 8.894386393810563e-05,
|
| 365 |
+
"loss": 0.0718,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.2,
|
| 370 |
+
"grad_norm": 0.577790379524231,
|
| 371 |
+
"learning_rate": 8.842005554284296e-05,
|
| 372 |
+
"loss": 0.0625,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 5.3,
|
| 377 |
+
"grad_norm": 0.7675192952156067,
|
| 378 |
+
"learning_rate": 8.788574348801675e-05,
|
| 379 |
+
"loss": 0.0637,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 5.4,
|
| 384 |
+
"grad_norm": 0.5569060444831848,
|
| 385 |
+
"learning_rate": 8.73410738492077e-05,
|
| 386 |
+
"loss": 0.0667,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 5.5,
|
| 391 |
+
"grad_norm": 0.6511795520782471,
|
| 392 |
+
"learning_rate": 8.678619553365659e-05,
|
| 393 |
+
"loss": 0.07,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 5.6,
|
| 398 |
+
"grad_norm": 0.7688615322113037,
|
| 399 |
+
"learning_rate": 8.622126023955446e-05,
|
| 400 |
+
"loss": 0.0579,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 5.7,
|
| 405 |
+
"grad_norm": 0.668289840221405,
|
| 406 |
+
"learning_rate": 8.564642241456986e-05,
|
| 407 |
+
"loss": 0.0599,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 5.8,
|
| 412 |
+
"grad_norm": 0.8834435939788818,
|
| 413 |
+
"learning_rate": 8.506183921362443e-05,
|
| 414 |
+
"loss": 0.0639,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 5.9,
|
| 419 |
+
"grad_norm": 0.727353572845459,
|
| 420 |
+
"learning_rate": 8.44676704559283e-05,
|
| 421 |
+
"loss": 0.0675,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.0,
|
| 426 |
+
"grad_norm": 2.630605936050415,
|
| 427 |
+
"learning_rate": 8.386407858128706e-05,
|
| 428 |
+
"loss": 0.0699,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.1,
|
| 433 |
+
"grad_norm": 1.2145522832870483,
|
| 434 |
+
"learning_rate": 8.32512286056924e-05,
|
| 435 |
+
"loss": 0.0724,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 6.2,
|
| 440 |
+
"grad_norm": 1.142543911933899,
|
| 441 |
+
"learning_rate": 8.262928807620843e-05,
|
| 442 |
+
"loss": 0.0727,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 6.3,
|
| 447 |
+
"grad_norm": 0.6511779427528381,
|
| 448 |
+
"learning_rate": 8.199842702516583e-05,
|
| 449 |
+
"loss": 0.0611,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 6.4,
|
| 454 |
+
"grad_norm": 0.5131811499595642,
|
| 455 |
+
"learning_rate": 8.135881792367686e-05,
|
| 456 |
+
"loss": 0.0591,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 6.5,
|
| 461 |
+
"grad_norm": 0.8611190915107727,
|
| 462 |
+
"learning_rate": 8.07106356344834e-05,
|
| 463 |
+
"loss": 0.0644,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 6.6,
|
| 468 |
+
"grad_norm": 0.7479195594787598,
|
| 469 |
+
"learning_rate": 8.005405736415126e-05,
|
| 470 |
+
"loss": 0.0638,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 6.7,
|
| 475 |
+
"grad_norm": 1.030855655670166,
|
| 476 |
+
"learning_rate": 7.938926261462366e-05,
|
| 477 |
+
"loss": 0.0644,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 6.8,
|
| 482 |
+
"grad_norm": 0.9318546056747437,
|
| 483 |
+
"learning_rate": 7.871643313414718e-05,
|
| 484 |
+
"loss": 0.0698,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 6.9,
|
| 489 |
+
"grad_norm": 0.8005062937736511,
|
| 490 |
+
"learning_rate": 7.803575286758364e-05,
|
| 491 |
+
"loss": 0.064,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.0,
|
| 496 |
+
"grad_norm": 1.5523598194122314,
|
| 497 |
+
"learning_rate": 7.734740790612136e-05,
|
| 498 |
+
"loss": 0.0581,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.1,
|
| 503 |
+
"grad_norm": 0.7943648099899292,
|
| 504 |
+
"learning_rate": 7.66515864363997e-05,
|
| 505 |
+
"loss": 0.0699,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.2,
|
| 510 |
+
"grad_norm": 0.775920569896698,
|
| 511 |
+
"learning_rate": 7.594847868906076e-05,
|
| 512 |
+
"loss": 0.0643,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.3,
|
| 517 |
+
"grad_norm": 0.7172490954399109,
|
| 518 |
+
"learning_rate": 7.52382768867422e-05,
|
| 519 |
+
"loss": 0.055,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.4,
|
| 524 |
+
"grad_norm": 0.7170266509056091,
|
| 525 |
+
"learning_rate": 7.452117519152542e-05,
|
| 526 |
+
"loss": 0.0561,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.5,
|
| 531 |
+
"grad_norm": 0.5932777523994446,
|
| 532 |
+
"learning_rate": 7.379736965185368e-05,
|
| 533 |
+
"loss": 0.0541,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.6,
|
| 538 |
+
"grad_norm": 0.5876290202140808,
|
| 539 |
+
"learning_rate": 7.30670581489344e-05,
|
| 540 |
+
"loss": 0.0568,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.7,
|
| 545 |
+
"grad_norm": 0.7758692502975464,
|
| 546 |
+
"learning_rate": 7.233044034264034e-05,
|
| 547 |
+
"loss": 0.055,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.8,
|
| 552 |
+
"grad_norm": 0.6511589288711548,
|
| 553 |
+
"learning_rate": 7.158771761692464e-05,
|
| 554 |
+
"loss": 0.0581,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.9,
|
| 559 |
+
"grad_norm": 0.777108371257782,
|
| 560 |
+
"learning_rate": 7.083909302476453e-05,
|
| 561 |
+
"loss": 0.0491,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 8.0,
|
| 566 |
+
"grad_norm": 3.6956264972686768,
|
| 567 |
+
"learning_rate": 7.008477123264848e-05,
|
| 568 |
+
"loss": 0.057,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 8.1,
|
| 573 |
+
"grad_norm": 0.69562828540802,
|
| 574 |
+
"learning_rate": 6.932495846462261e-05,
|
| 575 |
+
"loss": 0.0572,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 8.2,
|
| 580 |
+
"grad_norm": 0.7596244812011719,
|
| 581 |
+
"learning_rate": 6.855986244591104e-05,
|
| 582 |
+
"loss": 0.0598,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 8.3,
|
| 587 |
+
"grad_norm": 0.8167744278907776,
|
| 588 |
+
"learning_rate": 6.778969234612584e-05,
|
| 589 |
+
"loss": 0.0523,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 8.4,
|
| 594 |
+
"grad_norm": 0.7772257328033447,
|
| 595 |
+
"learning_rate": 6.701465872208216e-05,
|
| 596 |
+
"loss": 0.0563,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 8.5,
|
| 601 |
+
"grad_norm": 0.6254410147666931,
|
| 602 |
+
"learning_rate": 6.623497346023418e-05,
|
| 603 |
+
"loss": 0.0527,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 8.6,
|
| 608 |
+
"grad_norm": 0.4882695972919464,
|
| 609 |
+
"learning_rate": 6.545084971874738e-05,
|
| 610 |
+
"loss": 0.0543,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 8.7,
|
| 615 |
+
"grad_norm": 0.8698508739471436,
|
| 616 |
+
"learning_rate": 6.466250186922325e-05,
|
| 617 |
+
"loss": 0.0495,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 8.8,
|
| 622 |
+
"grad_norm": 0.6745582818984985,
|
| 623 |
+
"learning_rate": 6.387014543809223e-05,
|
| 624 |
+
"loss": 0.0505,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 8.9,
|
| 629 |
+
"grad_norm": 1.0861890316009521,
|
| 630 |
+
"learning_rate": 6.307399704769099e-05,
|
| 631 |
+
"loss": 0.044,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 9.0,
|
| 636 |
+
"grad_norm": 0.43595385551452637,
|
| 637 |
+
"learning_rate": 6.227427435703997e-05,
|
| 638 |
+
"loss": 0.0416,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 9.1,
|
| 643 |
+
"grad_norm": 1.1349962949752808,
|
| 644 |
+
"learning_rate": 6.147119600233758e-05,
|
| 645 |
+
"loss": 0.0576,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 9.2,
|
| 650 |
+
"grad_norm": 0.8424226641654968,
|
| 651 |
+
"learning_rate": 6.066498153718735e-05,
|
| 652 |
+
"loss": 0.0478,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 9.3,
|
| 657 |
+
"grad_norm": 1.0484619140625,
|
| 658 |
+
"learning_rate": 5.985585137257401e-05,
|
| 659 |
+
"loss": 0.0549,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 9.4,
|
| 664 |
+
"grad_norm": 0.6300575137138367,
|
| 665 |
+
"learning_rate": 5.90440267166055e-05,
|
| 666 |
+
"loss": 0.0507,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 9.5,
|
| 671 |
+
"grad_norm": 0.6857895851135254,
|
| 672 |
+
"learning_rate": 5.8229729514036705e-05,
|
| 673 |
+
"loss": 0.0463,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 9.6,
|
| 678 |
+
"grad_norm": 0.6959238648414612,
|
| 679 |
+
"learning_rate": 5.74131823855921e-05,
|
| 680 |
+
"loss": 0.0529,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 9.7,
|
| 685 |
+
"grad_norm": 0.6516974568367004,
|
| 686 |
+
"learning_rate": 5.6594608567103456e-05,
|
| 687 |
+
"loss": 0.0461,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 9.8,
|
| 692 |
+
"grad_norm": 1.1341180801391602,
|
| 693 |
+
"learning_rate": 5.577423184847932e-05,
|
| 694 |
+
"loss": 0.0458,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 9.9,
|
| 699 |
+
"grad_norm": 0.6836873292922974,
|
| 700 |
+
"learning_rate": 5.495227651252315e-05,
|
| 701 |
+
"loss": 0.0498,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 10.0,
|
| 706 |
+
"grad_norm": 1.8491439819335938,
|
| 707 |
+
"learning_rate": 5.4128967273616625e-05,
|
| 708 |
+
"loss": 0.0459,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 10.1,
|
| 713 |
+
"grad_norm": 0.7286385297775269,
|
| 714 |
+
"learning_rate": 5.330452921628497e-05,
|
| 715 |
+
"loss": 0.0495,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 10.2,
|
| 720 |
+
"grad_norm": 0.9026761651039124,
|
| 721 |
+
"learning_rate": 5.247918773366112e-05,
|
| 722 |
+
"loss": 0.0439,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 10.3,
|
| 727 |
+
"grad_norm": 0.5952501893043518,
|
| 728 |
+
"learning_rate": 5.165316846586541e-05,
|
| 729 |
+
"loss": 0.0438,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 10.4,
|
| 734 |
+
"grad_norm": 0.7074761986732483,
|
| 735 |
+
"learning_rate": 5.0826697238317935e-05,
|
| 736 |
+
"loss": 0.0444,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 10.5,
|
| 741 |
+
"grad_norm": 0.7149840593338013,
|
| 742 |
+
"learning_rate": 5e-05,
|
| 743 |
+
"loss": 0.0464,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 10.6,
|
| 748 |
+
"grad_norm": 0.9573127627372742,
|
| 749 |
+
"learning_rate": 4.917330276168208e-05,
|
| 750 |
+
"loss": 0.0434,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 10.7,
|
| 755 |
+
"grad_norm": 0.3760146498680115,
|
| 756 |
+
"learning_rate": 4.834683153413459e-05,
|
| 757 |
+
"loss": 0.0536,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 10.8,
|
| 762 |
+
"grad_norm": 0.7248881459236145,
|
| 763 |
+
"learning_rate": 4.7520812266338885e-05,
|
| 764 |
+
"loss": 0.052,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 10.9,
|
| 769 |
+
"grad_norm": 0.7167366147041321,
|
| 770 |
+
"learning_rate": 4.669547078371504e-05,
|
| 771 |
+
"loss": 0.0511,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 11.0,
|
| 776 |
+
"grad_norm": 0.554724931716919,
|
| 777 |
+
"learning_rate": 4.5871032726383386e-05,
|
| 778 |
+
"loss": 0.0361,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 11.1,
|
| 783 |
+
"grad_norm": 0.8976012468338013,
|
| 784 |
+
"learning_rate": 4.504772348747687e-05,
|
| 785 |
+
"loss": 0.0446,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 11.2,
|
| 790 |
+
"grad_norm": 1.2764275074005127,
|
| 791 |
+
"learning_rate": 4.4225768151520694e-05,
|
| 792 |
+
"loss": 0.0477,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 11.3,
|
| 797 |
+
"grad_norm": 0.9568806886672974,
|
| 798 |
+
"learning_rate": 4.3405391432896555e-05,
|
| 799 |
+
"loss": 0.0489,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 11.4,
|
| 804 |
+
"grad_norm": 0.5885640978813171,
|
| 805 |
+
"learning_rate": 4.2586817614407895e-05,
|
| 806 |
+
"loss": 0.0534,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 11.5,
|
| 811 |
+
"grad_norm": 0.48282381892204285,
|
| 812 |
+
"learning_rate": 4.17702704859633e-05,
|
| 813 |
+
"loss": 0.0489,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 11.6,
|
| 818 |
+
"grad_norm": 0.6082215905189514,
|
| 819 |
+
"learning_rate": 4.095597328339452e-05,
|
| 820 |
+
"loss": 0.0455,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 11.7,
|
| 825 |
+
"grad_norm": 0.8037371635437012,
|
| 826 |
+
"learning_rate": 4.0144148627425993e-05,
|
| 827 |
+
"loss": 0.0454,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 11.8,
|
| 832 |
+
"grad_norm": 0.6086834669113159,
|
| 833 |
+
"learning_rate": 3.933501846281267e-05,
|
| 834 |
+
"loss": 0.0427,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 11.9,
|
| 839 |
+
"grad_norm": 0.6202229857444763,
|
| 840 |
+
"learning_rate": 3.852880399766243e-05,
|
| 841 |
+
"loss": 0.0434,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 12.0,
|
| 846 |
+
"grad_norm": 1.6280860900878906,
|
| 847 |
+
"learning_rate": 3.772572564296005e-05,
|
| 848 |
+
"loss": 0.0405,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 12.1,
|
| 853 |
+
"grad_norm": 0.7294513583183289,
|
| 854 |
+
"learning_rate": 3.6926002952309016e-05,
|
| 855 |
+
"loss": 0.0474,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 12.2,
|
| 860 |
+
"grad_norm": 1.0804405212402344,
|
| 861 |
+
"learning_rate": 3.612985456190778e-05,
|
| 862 |
+
"loss": 0.0455,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 12.3,
|
| 867 |
+
"grad_norm": 0.8139125108718872,
|
| 868 |
+
"learning_rate": 3.533749813077677e-05,
|
| 869 |
+
"loss": 0.0396,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 12.4,
|
| 874 |
+
"grad_norm": 0.6082015037536621,
|
| 875 |
+
"learning_rate": 3.4549150281252636e-05,
|
| 876 |
+
"loss": 0.0455,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 12.5,
|
| 881 |
+
"grad_norm": 0.5969621539115906,
|
| 882 |
+
"learning_rate": 3.3765026539765834e-05,
|
| 883 |
+
"loss": 0.0393,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 12.6,
|
| 888 |
+
"grad_norm": 0.6664546728134155,
|
| 889 |
+
"learning_rate": 3.298534127791785e-05,
|
| 890 |
+
"loss": 0.0448,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 12.7,
|
| 895 |
+
"grad_norm": 0.5217233300209045,
|
| 896 |
+
"learning_rate": 3.221030765387417e-05,
|
| 897 |
+
"loss": 0.0446,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 12.8,
|
| 902 |
+
"grad_norm": 0.7529652118682861,
|
| 903 |
+
"learning_rate": 3.144013755408895e-05,
|
| 904 |
+
"loss": 0.0426,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 12.9,
|
| 909 |
+
"grad_norm": 0.41002699732780457,
|
| 910 |
+
"learning_rate": 3.0675041535377405e-05,
|
| 911 |
+
"loss": 0.0438,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 13.0,
|
| 916 |
+
"grad_norm": 3.2146363258361816,
|
| 917 |
+
"learning_rate": 2.991522876735154e-05,
|
| 918 |
+
"loss": 0.0435,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 13.1,
|
| 923 |
+
"grad_norm": 0.6749954223632812,
|
| 924 |
+
"learning_rate": 2.916090697523549e-05,
|
| 925 |
+
"loss": 0.0468,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 13.2,
|
| 930 |
+
"grad_norm": 0.7005327343940735,
|
| 931 |
+
"learning_rate": 2.8412282383075363e-05,
|
| 932 |
+
"loss": 0.0357,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 13.3,
|
| 937 |
+
"grad_norm": 0.6995335817337036,
|
| 938 |
+
"learning_rate": 2.766955965735968e-05,
|
| 939 |
+
"loss": 0.0426,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 13.4,
|
| 944 |
+
"grad_norm": 0.6105914115905762,
|
| 945 |
+
"learning_rate": 2.693294185106562e-05,
|
| 946 |
+
"loss": 0.0406,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 13.5,
|
| 951 |
+
"grad_norm": 0.6400395035743713,
|
| 952 |
+
"learning_rate": 2.6202630348146324e-05,
|
| 953 |
+
"loss": 0.043,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 13.6,
|
| 958 |
+
"grad_norm": 0.8495157957077026,
|
| 959 |
+
"learning_rate": 2.547882480847461e-05,
|
| 960 |
+
"loss": 0.0403,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 13.7,
|
| 965 |
+
"grad_norm": 0.6124421954154968,
|
| 966 |
+
"learning_rate": 2.476172311325783e-05,
|
| 967 |
+
"loss": 0.0395,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 13.8,
|
| 972 |
+
"grad_norm": 0.9979932904243469,
|
| 973 |
+
"learning_rate": 2.405152131093926e-05,
|
| 974 |
+
"loss": 0.0386,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 13.9,
|
| 979 |
+
"grad_norm": 0.9076727628707886,
|
| 980 |
+
"learning_rate": 2.3348413563600325e-05,
|
| 981 |
+
"loss": 0.0391,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 14.0,
|
| 986 |
+
"grad_norm": 0.8010100722312927,
|
| 987 |
+
"learning_rate": 2.2652592093878666e-05,
|
| 988 |
+
"loss": 0.0327,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 14.1,
|
| 993 |
+
"grad_norm": 0.7930116653442383,
|
| 994 |
+
"learning_rate": 2.196424713241637e-05,
|
| 995 |
+
"loss": 0.0416,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 14.2,
|
| 1000 |
+
"grad_norm": 0.4719603359699249,
|
| 1001 |
+
"learning_rate": 2.128356686585282e-05,
|
| 1002 |
+
"loss": 0.0339,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 14.3,
|
| 1007 |
+
"grad_norm": 0.5977267622947693,
|
| 1008 |
+
"learning_rate": 2.061073738537635e-05,
|
| 1009 |
+
"loss": 0.0433,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 14.4,
|
| 1014 |
+
"grad_norm": 0.924698531627655,
|
| 1015 |
+
"learning_rate": 1.9945942635848748e-05,
|
| 1016 |
+
"loss": 0.0436,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 14.5,
|
| 1021 |
+
"grad_norm": 0.3880954384803772,
|
| 1022 |
+
"learning_rate": 1.928936436551661e-05,
|
| 1023 |
+
"loss": 0.0403,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 14.6,
|
| 1028 |
+
"grad_norm": 0.4617973566055298,
|
| 1029 |
+
"learning_rate": 1.8641182076323148e-05,
|
| 1030 |
+
"loss": 0.0373,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 14.7,
|
| 1035 |
+
"grad_norm": 0.666461169719696,
|
| 1036 |
+
"learning_rate": 1.800157297483417e-05,
|
| 1037 |
+
"loss": 0.0418,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 14.8,
|
| 1042 |
+
"grad_norm": 0.4455702304840088,
|
| 1043 |
+
"learning_rate": 1.7370711923791567e-05,
|
| 1044 |
+
"loss": 0.0352,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 14.9,
|
| 1049 |
+
"grad_norm": 1.0331430435180664,
|
| 1050 |
+
"learning_rate": 1.6748771394307585e-05,
|
| 1051 |
+
"loss": 0.05,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 15.0,
|
| 1056 |
+
"grad_norm": 1.1350957155227661,
|
| 1057 |
+
"learning_rate": 1.6135921418712956e-05,
|
| 1058 |
+
"loss": 0.0365,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 15.1,
|
| 1063 |
+
"grad_norm": 0.6353791952133179,
|
| 1064 |
+
"learning_rate": 1.553232954407171e-05,
|
| 1065 |
+
"loss": 0.0399,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 15.2,
|
| 1070 |
+
"grad_norm": 0.5416865944862366,
|
| 1071 |
+
"learning_rate": 1.4938160786375572e-05,
|
| 1072 |
+
"loss": 0.0451,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 15.3,
|
| 1077 |
+
"grad_norm": 0.864016056060791,
|
| 1078 |
+
"learning_rate": 1.435357758543015e-05,
|
| 1079 |
+
"loss": 0.0379,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 15.4,
|
| 1084 |
+
"grad_norm": 0.4366832971572876,
|
| 1085 |
+
"learning_rate": 1.3778739760445552e-05,
|
| 1086 |
+
"loss": 0.0405,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 15.5,
|
| 1091 |
+
"grad_norm": 0.5444951057434082,
|
| 1092 |
+
"learning_rate": 1.3213804466343421e-05,
|
| 1093 |
+
"loss": 0.0368,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 15.6,
|
| 1098 |
+
"grad_norm": 1.8182613849639893,
|
| 1099 |
+
"learning_rate": 1.2658926150792322e-05,
|
| 1100 |
+
"loss": 0.0497,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 15.7,
|
| 1105 |
+
"grad_norm": 0.5947335362434387,
|
| 1106 |
+
"learning_rate": 1.2114256511983274e-05,
|
| 1107 |
+
"loss": 0.0391,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 15.8,
|
| 1112 |
+
"grad_norm": 0.794297993183136,
|
| 1113 |
+
"learning_rate": 1.157994445715706e-05,
|
| 1114 |
+
"loss": 0.0372,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 15.9,
|
| 1119 |
+
"grad_norm": 0.6530230641365051,
|
| 1120 |
+
"learning_rate": 1.1056136061894384e-05,
|
| 1121 |
+
"loss": 0.0385,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 16.0,
|
| 1126 |
+
"grad_norm": 2.78792405128479,
|
| 1127 |
+
"learning_rate": 1.0542974530180327e-05,
|
| 1128 |
+
"loss": 0.0485,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 16.1,
|
| 1133 |
+
"grad_norm": 0.9636825323104858,
|
| 1134 |
+
"learning_rate": 1.0040600155253765e-05,
|
| 1135 |
+
"loss": 0.0371,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 16.2,
|
| 1140 |
+
"grad_norm": 0.4774838984012604,
|
| 1141 |
+
"learning_rate": 9.549150281252633e-06,
|
| 1142 |
+
"loss": 0.0379,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 16.3,
|
| 1147 |
+
"grad_norm": 0.6014675498008728,
|
| 1148 |
+
"learning_rate": 9.068759265665384e-06,
|
| 1149 |
+
"loss": 0.0354,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 16.4,
|
| 1154 |
+
"grad_norm": 0.7133916020393372,
|
| 1155 |
+
"learning_rate": 8.599558442598998e-06,
|
| 1156 |
+
"loss": 0.036,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 16.5,
|
| 1161 |
+
"grad_norm": 0.765894889831543,
|
| 1162 |
+
"learning_rate": 8.141676086873572e-06,
|
| 1163 |
+
"loss": 0.0371,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 16.6,
|
| 1168 |
+
"grad_norm": 0.49927645921707153,
|
| 1169 |
+
"learning_rate": 7.695237378953223e-06,
|
| 1170 |
+
"loss": 0.0398,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 16.7,
|
| 1175 |
+
"grad_norm": 1.3441693782806396,
|
| 1176 |
+
"learning_rate": 7.260364370723044e-06,
|
| 1177 |
+
"loss": 0.0393,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 16.8,
|
| 1182 |
+
"grad_norm": 0.6928268671035767,
|
| 1183 |
+
"learning_rate": 6.837175952121306e-06,
|
| 1184 |
+
"loss": 0.0333,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 16.9,
|
| 1189 |
+
"grad_norm": 0.4891086220741272,
|
| 1190 |
+
"learning_rate": 6.425787818636131e-06,
|
| 1191 |
+
"loss": 0.0357,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 17.0,
|
| 1196 |
+
"grad_norm": 1.0727057456970215,
|
| 1197 |
+
"learning_rate": 6.026312439675552e-06,
|
| 1198 |
+
"loss": 0.0403,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 17.1,
|
| 1203 |
+
"grad_norm": 0.5739259123802185,
|
| 1204 |
+
"learning_rate": 5.6388590278194096e-06,
|
| 1205 |
+
"loss": 0.0366,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 17.2,
|
| 1210 |
+
"grad_norm": 0.5994676351547241,
|
| 1211 |
+
"learning_rate": 5.263533508961827e-06,
|
| 1212 |
+
"loss": 0.0365,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 17.3,
|
| 1217 |
+
"grad_norm": 0.7043574452400208,
|
| 1218 |
+
"learning_rate": 4.900438493352055e-06,
|
| 1219 |
+
"loss": 0.0404,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 17.4,
|
| 1224 |
+
"grad_norm": 0.4768543839454651,
|
| 1225 |
+
"learning_rate": 4.549673247541875e-06,
|
| 1226 |
+
"loss": 0.0419,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 17.5,
|
| 1231 |
+
"grad_norm": 0.800778329372406,
|
| 1232 |
+
"learning_rate": 4.2113336672471245e-06,
|
| 1233 |
+
"loss": 0.0379,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 17.6,
|
| 1238 |
+
"grad_norm": 0.9242330193519592,
|
| 1239 |
+
"learning_rate": 3.885512251130763e-06,
|
| 1240 |
+
"loss": 0.0337,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 17.7,
|
| 1245 |
+
"grad_norm": 0.37277647852897644,
|
| 1246 |
+
"learning_rate": 3.5722980755146517e-06,
|
| 1247 |
+
"loss": 0.0381,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 17.8,
|
| 1252 |
+
"grad_norm": 0.8925457000732422,
|
| 1253 |
+
"learning_rate": 3.271776770026963e-06,
|
| 1254 |
+
"loss": 0.0383,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 17.9,
|
| 1259 |
+
"grad_norm": 0.8878549337387085,
|
| 1260 |
+
"learning_rate": 2.9840304941919415e-06,
|
| 1261 |
+
"loss": 0.0405,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 18.0,
|
| 1266 |
+
"grad_norm": 0.779109537601471,
|
| 1267 |
+
"learning_rate": 2.7091379149682685e-06,
|
| 1268 |
+
"loss": 0.0341,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 18.1,
|
| 1273 |
+
"grad_norm": 0.4987022280693054,
|
| 1274 |
+
"learning_rate": 2.4471741852423237e-06,
|
| 1275 |
+
"loss": 0.0435,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 18.2,
|
| 1280 |
+
"grad_norm": 0.8446706533432007,
|
| 1281 |
+
"learning_rate": 2.1982109232821178e-06,
|
| 1282 |
+
"loss": 0.0401,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 18.3,
|
| 1287 |
+
"grad_norm": 0.6136398911476135,
|
| 1288 |
+
"learning_rate": 1.962316193157593e-06,
|
| 1289 |
+
"loss": 0.0405,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 18.4,
|
| 1294 |
+
"grad_norm": 0.5189019441604614,
|
| 1295 |
+
"learning_rate": 1.7395544861325718e-06,
|
| 1296 |
+
"loss": 0.035,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 18.5,
|
| 1301 |
+
"grad_norm": 1.2081691026687622,
|
| 1302 |
+
"learning_rate": 1.5299867030334814e-06,
|
| 1303 |
+
"loss": 0.0301,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 18.6,
|
| 1308 |
+
"grad_norm": 0.5128328204154968,
|
| 1309 |
+
"learning_rate": 1.333670137599713e-06,
|
| 1310 |
+
"loss": 0.0342,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 18.7,
|
| 1315 |
+
"grad_norm": 0.6271176338195801,
|
| 1316 |
+
"learning_rate": 1.1506584608200367e-06,
|
| 1317 |
+
"loss": 0.0374,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 18.8,
|
| 1322 |
+
"grad_norm": 0.7536263465881348,
|
| 1323 |
+
"learning_rate": 9.810017062595322e-07,
|
| 1324 |
+
"loss": 0.0432,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 18.9,
|
| 1329 |
+
"grad_norm": 0.70732581615448,
|
| 1330 |
+
"learning_rate": 8.247462563808817e-07,
|
| 1331 |
+
"loss": 0.0389,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 19.0,
|
| 1336 |
+
"grad_norm": 3.046426296234131,
|
| 1337 |
+
"learning_rate": 6.819348298638839e-07,
|
| 1338 |
+
"loss": 0.0339,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 19.1,
|
| 1343 |
+
"grad_norm": 0.5528438091278076,
|
| 1344 |
+
"learning_rate": 5.526064699265753e-07,
|
| 1345 |
+
"loss": 0.0356,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 19.2,
|
| 1350 |
+
"grad_norm": 0.4317626357078552,
|
| 1351 |
+
"learning_rate": 4.367965336512403e-07,
|
| 1352 |
+
"loss": 0.0415,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 19.3,
|
| 1357 |
+
"grad_norm": 0.7551414966583252,
|
| 1358 |
+
"learning_rate": 3.3453668231809286e-07,
|
| 1359 |
+
"loss": 0.0375,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 19.4,
|
| 1364 |
+
"grad_norm": 0.737501859664917,
|
| 1365 |
+
"learning_rate": 2.458548727494292e-07,
|
| 1366 |
+
"loss": 0.0391,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 19.5,
|
| 1371 |
+
"grad_norm": 0.3441050052642822,
|
| 1372 |
+
"learning_rate": 1.7077534966650766e-07,
|
| 1373 |
+
"loss": 0.0356,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 19.6,
|
| 1378 |
+
"grad_norm": 0.7213258147239685,
|
| 1379 |
+
"learning_rate": 1.0931863906127327e-07,
|
| 1380 |
+
"loss": 0.0377,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 19.7,
|
| 1385 |
+
"grad_norm": 0.6739200353622437,
|
| 1386 |
+
"learning_rate": 6.150154258476315e-08,
|
| 1387 |
+
"loss": 0.0346,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 19.8,
|
| 1392 |
+
"grad_norm": 0.4855494499206543,
|
| 1393 |
+
"learning_rate": 2.7337132953697554e-08,
|
| 1394 |
+
"loss": 0.0315,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 19.9,
|
| 1399 |
+
"grad_norm": 0.8237863779067993,
|
| 1400 |
+
"learning_rate": 6.834750376549792e-09,
|
| 1401 |
+
"loss": 0.0403,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 20.0,
|
| 1406 |
+
"grad_norm": 0.4662252366542816,
|
| 1407 |
+
"learning_rate": 0.0,
|
| 1408 |
+
"loss": 0.0408,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 20.0,
|
| 1413 |
+
"step": 2000,
|
| 1414 |
+
"total_flos": 1.357860048774336e+17,
|
| 1415 |
+
"train_loss": 0.09518647351861,
|
| 1416 |
+
"train_runtime": 632.3751,
|
| 1417 |
+
"train_samples_per_second": 100.257,
|
| 1418 |
+
"train_steps_per_second": 3.163
|
| 1419 |
+
}
|
| 1420 |
+
],
|
| 1421 |
+
"logging_steps": 10,
|
| 1422 |
+
"max_steps": 2000,
|
| 1423 |
+
"num_input_tokens_seen": 0,
|
| 1424 |
+
"num_train_epochs": 20,
|
| 1425 |
+
"save_steps": 20000,
|
| 1426 |
+
"stateful_callbacks": {
|
| 1427 |
+
"TrainerControl": {
|
| 1428 |
+
"args": {
|
| 1429 |
+
"should_epoch_stop": false,
|
| 1430 |
+
"should_evaluate": false,
|
| 1431 |
+
"should_log": false,
|
| 1432 |
+
"should_save": true,
|
| 1433 |
+
"should_training_stop": true
|
| 1434 |
+
},
|
| 1435 |
+
"attributes": {}
|
| 1436 |
+
}
|
| 1437 |
+
},
|
| 1438 |
+
"total_flos": 1.357860048774336e+17,
|
| 1439 |
+
"train_batch_size": 32,
|
| 1440 |
+
"trial_name": null,
|
| 1441 |
+
"trial_params": null
|
| 1442 |
+
}
|