Fathom: upload expert-e5-threatintel/training_log.json
Browse files
adapters/expert-e5-threatintel/training_log.json
ADDED
|
@@ -0,0 +1,1501 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 1.5327,
|
| 4 |
+
"grad_norm": 0.5706859827041626,
|
| 5 |
+
"learning_rate": 4.5e-05,
|
| 6 |
+
"entropy": 1.2147093176841737,
|
| 7 |
+
"num_tokens": 156309.0,
|
| 8 |
+
"mean_token_accuracy": 0.659215685725212,
|
| 9 |
+
"epoch": 0.03355704697986577,
|
| 10 |
+
"step": 10
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"loss": 1.1979,
|
| 14 |
+
"grad_norm": 1.02685546875,
|
| 15 |
+
"learning_rate": 9.5e-05,
|
| 16 |
+
"entropy": 1.2502837777137756,
|
| 17 |
+
"num_tokens": 307512.0,
|
| 18 |
+
"mean_token_accuracy": 0.7190202981233597,
|
| 19 |
+
"epoch": 0.06711409395973154,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"loss": 0.6687,
|
| 24 |
+
"grad_norm": 0.5571576356887817,
|
| 25 |
+
"learning_rate": 9.999075138471951e-05,
|
| 26 |
+
"entropy": 0.6448976010084152,
|
| 27 |
+
"num_tokens": 461226.0,
|
| 28 |
+
"mean_token_accuracy": 0.8424648404121399,
|
| 29 |
+
"epoch": 0.10067114093959731,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"loss": 0.4812,
|
| 34 |
+
"grad_norm": 0.38987967371940613,
|
| 35 |
+
"learning_rate": 9.995878525539525e-05,
|
| 36 |
+
"entropy": 0.47405652403831483,
|
| 37 |
+
"num_tokens": 618054.0,
|
| 38 |
+
"mean_token_accuracy": 0.8851036787033081,
|
| 39 |
+
"epoch": 0.1342281879194631,
|
| 40 |
+
"step": 40
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"loss": 0.3701,
|
| 44 |
+
"grad_norm": 0.2643287777900696,
|
| 45 |
+
"learning_rate": 9.990400202763563e-05,
|
| 46 |
+
"entropy": 0.3693989284336567,
|
| 47 |
+
"num_tokens": 775209.0,
|
| 48 |
+
"mean_token_accuracy": 0.9097828209400177,
|
| 49 |
+
"epoch": 0.16778523489932887,
|
| 50 |
+
"step": 50
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"loss": 0.3193,
|
| 54 |
+
"grad_norm": 0.27373382449150085,
|
| 55 |
+
"learning_rate": 9.982642672195092e-05,
|
| 56 |
+
"entropy": 0.32156657576560976,
|
| 57 |
+
"num_tokens": 931365.0,
|
| 58 |
+
"mean_token_accuracy": 0.920613020658493,
|
| 59 |
+
"epoch": 0.20134228187919462,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"loss": 0.2866,
|
| 64 |
+
"grad_norm": 0.3094562292098999,
|
| 65 |
+
"learning_rate": 9.972609476841367e-05,
|
| 66 |
+
"entropy": 0.292499927431345,
|
| 67 |
+
"num_tokens": 1086695.0,
|
| 68 |
+
"mean_token_accuracy": 0.9282213032245636,
|
| 69 |
+
"epoch": 0.2348993288590604,
|
| 70 |
+
"step": 70
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"loss": 0.2489,
|
| 74 |
+
"grad_norm": 0.2749999761581421,
|
| 75 |
+
"learning_rate": 9.960305199047712e-05,
|
| 76 |
+
"entropy": 0.2526983417570591,
|
| 77 |
+
"num_tokens": 1241470.0,
|
| 78 |
+
"mean_token_accuracy": 0.9362666577100753,
|
| 79 |
+
"epoch": 0.2684563758389262,
|
| 80 |
+
"step": 80
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"loss": 0.2459,
|
| 84 |
+
"grad_norm": 0.2721326947212219,
|
| 85 |
+
"learning_rate": 9.945735458404681e-05,
|
| 86 |
+
"entropy": 0.24655402898788453,
|
| 87 |
+
"num_tokens": 1396709.0,
|
| 88 |
+
"mean_token_accuracy": 0.9367562144994735,
|
| 89 |
+
"epoch": 0.30201342281879195,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"loss": 0.2435,
|
| 94 |
+
"grad_norm": 0.21973158419132233,
|
| 95 |
+
"learning_rate": 9.928906909181481e-05,
|
| 96 |
+
"entropy": 0.24917888268828392,
|
| 97 |
+
"num_tokens": 1555025.0,
|
| 98 |
+
"mean_token_accuracy": 0.9364570289850235,
|
| 99 |
+
"epoch": 0.33557046979865773,
|
| 100 |
+
"step": 100
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"loss": 0.2347,
|
| 104 |
+
"grad_norm": 0.26128071546554565,
|
| 105 |
+
"learning_rate": 9.909827237286849e-05,
|
| 106 |
+
"entropy": 0.23597020357847215,
|
| 107 |
+
"num_tokens": 1706763.0,
|
| 108 |
+
"mean_token_accuracy": 0.9403788715600967,
|
| 109 |
+
"epoch": 0.3691275167785235,
|
| 110 |
+
"step": 110
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"loss": 0.2478,
|
| 114 |
+
"grad_norm": 0.20566503703594208,
|
| 115 |
+
"learning_rate": 9.888505156758759e-05,
|
| 116 |
+
"entropy": 0.2498910292983055,
|
| 117 |
+
"num_tokens": 1867753.0,
|
| 118 |
+
"mean_token_accuracy": 0.9367677628993988,
|
| 119 |
+
"epoch": 0.40268456375838924,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"loss": 0.2417,
|
| 124 |
+
"grad_norm": 0.190927654504776,
|
| 125 |
+
"learning_rate": 9.864950405784551e-05,
|
| 126 |
+
"entropy": 0.24509735256433487,
|
| 127 |
+
"num_tokens": 2027436.0,
|
| 128 |
+
"mean_token_accuracy": 0.9376339435577392,
|
| 129 |
+
"epoch": 0.436241610738255,
|
| 130 |
+
"step": 130
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"loss": 0.2196,
|
| 134 |
+
"grad_norm": 0.1996077448129654,
|
| 135 |
+
"learning_rate": 9.839173742253334e-05,
|
| 136 |
+
"entropy": 0.22244628816843032,
|
| 137 |
+
"num_tokens": 2182978.0,
|
| 138 |
+
"mean_token_accuracy": 0.9424097180366516,
|
| 139 |
+
"epoch": 0.4697986577181208,
|
| 140 |
+
"step": 140
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"loss": 0.2534,
|
| 144 |
+
"grad_norm": 0.21034187078475952,
|
| 145 |
+
"learning_rate": 9.811186938842645e-05,
|
| 146 |
+
"entropy": 0.2574092894792557,
|
| 147 |
+
"num_tokens": 2343108.0,
|
| 148 |
+
"mean_token_accuracy": 0.9341461777687072,
|
| 149 |
+
"epoch": 0.5033557046979866,
|
| 150 |
+
"step": 150
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"loss": 0.2369,
|
| 154 |
+
"grad_norm": 0.20116521418094635,
|
| 155 |
+
"learning_rate": 9.781002777641664e-05,
|
| 156 |
+
"entropy": 0.23710015863180162,
|
| 157 |
+
"num_tokens": 2503180.0,
|
| 158 |
+
"mean_token_accuracy": 0.9385264277458191,
|
| 159 |
+
"epoch": 0.5369127516778524,
|
| 160 |
+
"step": 160
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"loss": 0.2194,
|
| 164 |
+
"grad_norm": 0.20917288959026337,
|
| 165 |
+
"learning_rate": 9.748635044313386e-05,
|
| 166 |
+
"entropy": 0.22300637513399124,
|
| 167 |
+
"num_tokens": 2658840.0,
|
| 168 |
+
"mean_token_accuracy": 0.9410306721925735,
|
| 169 |
+
"epoch": 0.5704697986577181,
|
| 170 |
+
"step": 170
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"loss": 0.2295,
|
| 174 |
+
"grad_norm": 0.19810253381729126,
|
| 175 |
+
"learning_rate": 9.714098521798465e-05,
|
| 176 |
+
"entropy": 0.23066828176379203,
|
| 177 |
+
"num_tokens": 2818656.0,
|
| 178 |
+
"mean_token_accuracy": 0.9388932436704636,
|
| 179 |
+
"epoch": 0.6040268456375839,
|
| 180 |
+
"step": 180
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"loss": 0.231,
|
| 184 |
+
"grad_norm": 0.17845278978347778,
|
| 185 |
+
"learning_rate": 9.677408983563565e-05,
|
| 186 |
+
"entropy": 0.23543039262294768,
|
| 187 |
+
"num_tokens": 2975473.0,
|
| 188 |
+
"mean_token_accuracy": 0.9391510993242264,
|
| 189 |
+
"epoch": 0.6375838926174496,
|
| 190 |
+
"step": 190
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"loss": 0.2152,
|
| 194 |
+
"grad_norm": 0.23437312245368958,
|
| 195 |
+
"learning_rate": 9.638583186397331e-05,
|
| 196 |
+
"entropy": 0.21993101984262467,
|
| 197 |
+
"num_tokens": 3130452.0,
|
| 198 |
+
"mean_token_accuracy": 0.9423492521047592,
|
| 199 |
+
"epoch": 0.6711409395973155,
|
| 200 |
+
"step": 200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"loss": 0.204,
|
| 204 |
+
"grad_norm": 0.2377660572528839,
|
| 205 |
+
"learning_rate": 9.597638862757255e-05,
|
| 206 |
+
"entropy": 0.2064872995018959,
|
| 207 |
+
"num_tokens": 3287490.0,
|
| 208 |
+
"mean_token_accuracy": 0.9451317518949509,
|
| 209 |
+
"epoch": 0.7046979865771812,
|
| 210 |
+
"step": 210
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"loss": 0.2158,
|
| 214 |
+
"grad_norm": 0.17862877249717712,
|
| 215 |
+
"learning_rate": 9.554594712670926e-05,
|
| 216 |
+
"entropy": 0.2166791968047619,
|
| 217 |
+
"num_tokens": 3446266.0,
|
| 218 |
+
"mean_token_accuracy": 0.9421544283628464,
|
| 219 |
+
"epoch": 0.738255033557047,
|
| 220 |
+
"step": 220
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"loss": 0.2253,
|
| 224 |
+
"grad_norm": 0.1511947214603424,
|
| 225 |
+
"learning_rate": 9.509470395195399e-05,
|
| 226 |
+
"entropy": 0.22676953300833702,
|
| 227 |
+
"num_tokens": 3602351.0,
|
| 228 |
+
"mean_token_accuracy": 0.939826962351799,
|
| 229 |
+
"epoch": 0.7718120805369127,
|
| 230 |
+
"step": 230
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"loss": 0.2201,
|
| 234 |
+
"grad_norm": 0.26649826765060425,
|
| 235 |
+
"learning_rate": 9.46228651943853e-05,
|
| 236 |
+
"entropy": 0.22365730181336402,
|
| 237 |
+
"num_tokens": 3757932.0,
|
| 238 |
+
"mean_token_accuracy": 0.9410541921854019,
|
| 239 |
+
"epoch": 0.8053691275167785,
|
| 240 |
+
"step": 240
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"loss": 0.2026,
|
| 244 |
+
"grad_norm": 0.19634084403514862,
|
| 245 |
+
"learning_rate": 9.413064635146418e-05,
|
| 246 |
+
"entropy": 0.20426617376506329,
|
| 247 |
+
"num_tokens": 3912346.0,
|
| 248 |
+
"mean_token_accuracy": 0.9445015460252761,
|
| 249 |
+
"epoch": 0.8389261744966443,
|
| 250 |
+
"step": 250
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"loss": 0.2201,
|
| 254 |
+
"grad_norm": 0.19683024287223816,
|
| 255 |
+
"learning_rate": 9.361827222861241e-05,
|
| 256 |
+
"entropy": 0.22703441381454467,
|
| 257 |
+
"num_tokens": 4071117.0,
|
| 258 |
+
"mean_token_accuracy": 0.9421917259693146,
|
| 259 |
+
"epoch": 0.87248322147651,
|
| 260 |
+
"step": 260
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"loss": 0.2048,
|
| 264 |
+
"grad_norm": 0.184284046292305,
|
| 265 |
+
"learning_rate": 9.308597683653975e-05,
|
| 266 |
+
"entropy": 0.20578452795743943,
|
| 267 |
+
"num_tokens": 4226750.0,
|
| 268 |
+
"mean_token_accuracy": 0.9451743960380554,
|
| 269 |
+
"epoch": 0.9060402684563759,
|
| 270 |
+
"step": 270
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"loss": 0.2181,
|
| 274 |
+
"grad_norm": 0.19832965731620789,
|
| 275 |
+
"learning_rate": 9.253400328436699e-05,
|
| 276 |
+
"entropy": 0.22092150747776032,
|
| 277 |
+
"num_tokens": 4385096.0,
|
| 278 |
+
"mean_token_accuracy": 0.9426127344369888,
|
| 279 |
+
"epoch": 0.9395973154362416,
|
| 280 |
+
"step": 280
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"loss": 0.2178,
|
| 284 |
+
"grad_norm": 0.16237328946590424,
|
| 285 |
+
"learning_rate": 9.196260366859342e-05,
|
| 286 |
+
"entropy": 0.22047539427876472,
|
| 287 |
+
"num_tokens": 4544552.0,
|
| 288 |
+
"mean_token_accuracy": 0.9419336885213851,
|
| 289 |
+
"epoch": 0.9731543624161074,
|
| 290 |
+
"step": 290
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"loss": 0.2188,
|
| 294 |
+
"grad_norm": 0.19606098532676697,
|
| 295 |
+
"learning_rate": 9.137203895795983e-05,
|
| 296 |
+
"entropy": 0.2207832932472229,
|
| 297 |
+
"num_tokens": 4701426.0,
|
| 298 |
+
"mean_token_accuracy": 0.941154745221138,
|
| 299 |
+
"epoch": 1.0067114093959733,
|
| 300 |
+
"step": 300
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"loss": 0.1948,
|
| 304 |
+
"grad_norm": 0.2281929850578308,
|
| 305 |
+
"learning_rate": 9.076257887425923e-05,
|
| 306 |
+
"entropy": 0.19738901741802692,
|
| 307 |
+
"num_tokens": 4858846.0,
|
| 308 |
+
"mean_token_accuracy": 0.9471817642450333,
|
| 309 |
+
"epoch": 1.0402684563758389,
|
| 310 |
+
"step": 310
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"loss": 0.2037,
|
| 314 |
+
"grad_norm": 0.19096426665782928,
|
| 315 |
+
"learning_rate": 9.01345017691499e-05,
|
| 316 |
+
"entropy": 0.2047410562634468,
|
| 317 |
+
"num_tokens": 5015590.0,
|
| 318 |
+
"mean_token_accuracy": 0.9454032570123673,
|
| 319 |
+
"epoch": 1.0738255033557047,
|
| 320 |
+
"step": 320
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"loss": 0.2023,
|
| 324 |
+
"grad_norm": 0.19078002870082855,
|
| 325 |
+
"learning_rate": 8.948809449702711e-05,
|
| 326 |
+
"entropy": 0.21062878221273423,
|
| 327 |
+
"num_tokens": 5173149.0,
|
| 328 |
+
"mean_token_accuracy": 0.9446896910667419,
|
| 329 |
+
"epoch": 1.1073825503355705,
|
| 330 |
+
"step": 330
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"loss": 0.1997,
|
| 334 |
+
"grad_norm": 0.26884037256240845,
|
| 335 |
+
"learning_rate": 8.882365228401139e-05,
|
| 336 |
+
"entropy": 0.20285834297537803,
|
| 337 |
+
"num_tokens": 5331307.0,
|
| 338 |
+
"mean_token_accuracy": 0.9449720442295074,
|
| 339 |
+
"epoch": 1.1409395973154361,
|
| 340 |
+
"step": 340
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"loss": 0.1967,
|
| 344 |
+
"grad_norm": 0.23782381415367126,
|
| 345 |
+
"learning_rate": 8.814147859311332e-05,
|
| 346 |
+
"entropy": 0.1966269500553608,
|
| 347 |
+
"num_tokens": 5486558.0,
|
| 348 |
+
"mean_token_accuracy": 0.9466675192117691,
|
| 349 |
+
"epoch": 1.174496644295302,
|
| 350 |
+
"step": 350
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"loss": 0.2135,
|
| 354 |
+
"grad_norm": 0.19140051305294037,
|
| 355 |
+
"learning_rate": 8.744188498563641e-05,
|
| 356 |
+
"entropy": 0.21856041625142097,
|
| 357 |
+
"num_tokens": 5643311.0,
|
| 358 |
+
"mean_token_accuracy": 0.9413786977529526,
|
| 359 |
+
"epoch": 1.2080536912751678,
|
| 360 |
+
"step": 360
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"loss": 0.1956,
|
| 364 |
+
"grad_norm": 0.2523305118083954,
|
| 365 |
+
"learning_rate": 8.672519097888126e-05,
|
| 366 |
+
"entropy": 0.19831475540995597,
|
| 367 |
+
"num_tokens": 5798511.0,
|
| 368 |
+
"mean_token_accuracy": 0.946366423368454,
|
| 369 |
+
"epoch": 1.2416107382550337,
|
| 370 |
+
"step": 370
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"loss": 0.1895,
|
| 374 |
+
"grad_norm": 0.16881771385669708,
|
| 375 |
+
"learning_rate": 8.599172390021615e-05,
|
| 376 |
+
"entropy": 0.1935427539050579,
|
| 377 |
+
"num_tokens": 5954868.0,
|
| 378 |
+
"mean_token_accuracy": 0.9482783049345016,
|
| 379 |
+
"epoch": 1.2751677852348993,
|
| 380 |
+
"step": 380
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"loss": 0.1857,
|
| 384 |
+
"grad_norm": 0.15226209163665771,
|
| 385 |
+
"learning_rate": 8.524181873758059e-05,
|
| 386 |
+
"entropy": 0.18582973293960095,
|
| 387 |
+
"num_tokens": 6108894.0,
|
| 388 |
+
"mean_token_accuracy": 0.9488643199205399,
|
| 389 |
+
"epoch": 1.308724832214765,
|
| 390 |
+
"step": 390
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.2078,
|
| 394 |
+
"grad_norm": 0.19812646508216858,
|
| 395 |
+
"learning_rate": 8.447581798649014e-05,
|
| 396 |
+
"entropy": 0.20552278831601142,
|
| 397 |
+
"num_tokens": 6267764.0,
|
| 398 |
+
"mean_token_accuracy": 0.9442088425159454,
|
| 399 |
+
"epoch": 1.342281879194631,
|
| 400 |
+
"step": 400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"loss": 0.2074,
|
| 404 |
+
"grad_norm": 0.188148632645607,
|
| 405 |
+
"learning_rate": 8.369407149361241e-05,
|
| 406 |
+
"entropy": 0.21127415373921393,
|
| 407 |
+
"num_tokens": 6424716.0,
|
| 408 |
+
"mean_token_accuracy": 0.9449205726385117,
|
| 409 |
+
"epoch": 1.3758389261744965,
|
| 410 |
+
"step": 410
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"loss": 0.1948,
|
| 414 |
+
"grad_norm": 0.19974718987941742,
|
| 415 |
+
"learning_rate": 8.289693629698564e-05,
|
| 416 |
+
"entropy": 0.19758303910493852,
|
| 417 |
+
"num_tokens": 6579760.0,
|
| 418 |
+
"mean_token_accuracy": 0.9469632744789124,
|
| 419 |
+
"epoch": 1.4093959731543624,
|
| 420 |
+
"step": 420
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"loss": 0.2144,
|
| 424 |
+
"grad_norm": 0.2088315635919571,
|
| 425 |
+
"learning_rate": 8.208477646295277e-05,
|
| 426 |
+
"entropy": 0.21340604722499848,
|
| 427 |
+
"num_tokens": 6737911.0,
|
| 428 |
+
"mean_token_accuracy": 0.9427065312862396,
|
| 429 |
+
"epoch": 1.4429530201342282,
|
| 430 |
+
"step": 430
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"loss": 0.2145,
|
| 434 |
+
"grad_norm": 0.31156837940216064,
|
| 435 |
+
"learning_rate": 8.125796291988577e-05,
|
| 436 |
+
"entropy": 0.21349589377641678,
|
| 437 |
+
"num_tokens": 6894903.0,
|
| 438 |
+
"mean_token_accuracy": 0.9425385951995849,
|
| 439 |
+
"epoch": 1.476510067114094,
|
| 440 |
+
"step": 440
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"loss": 0.2159,
|
| 444 |
+
"grad_norm": 0.18016791343688965,
|
| 445 |
+
"learning_rate": 8.041687328877567e-05,
|
| 446 |
+
"entropy": 0.2204777255654335,
|
| 447 |
+
"num_tokens": 7054943.0,
|
| 448 |
+
"mean_token_accuracy": 0.9415547668933868,
|
| 449 |
+
"epoch": 1.5100671140939599,
|
| 450 |
+
"step": 450
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"loss": 0.202,
|
| 454 |
+
"grad_norm": 0.2058987319469452,
|
| 455 |
+
"learning_rate": 7.956189171076616e-05,
|
| 456 |
+
"entropy": 0.20117317102849483,
|
| 457 |
+
"num_tokens": 7211170.0,
|
| 458 |
+
"mean_token_accuracy": 0.9456082373857498,
|
| 459 |
+
"epoch": 1.5436241610738255,
|
| 460 |
+
"step": 460
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"loss": 0.1939,
|
| 464 |
+
"grad_norm": 0.15302836894989014,
|
| 465 |
+
"learning_rate": 7.869340867170928e-05,
|
| 466 |
+
"entropy": 0.19479563012719153,
|
| 467 |
+
"num_tokens": 7369121.0,
|
| 468 |
+
"mean_token_accuracy": 0.9471449553966522,
|
| 469 |
+
"epoch": 1.5771812080536913,
|
| 470 |
+
"step": 470
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"loss": 0.1886,
|
| 474 |
+
"grad_norm": 0.16370812058448792,
|
| 475 |
+
"learning_rate": 7.781182082382325e-05,
|
| 476 |
+
"entropy": 0.18975077904760837,
|
| 477 |
+
"num_tokens": 7525693.0,
|
| 478 |
+
"mean_token_accuracy": 0.9485057532787323,
|
| 479 |
+
"epoch": 1.610738255033557,
|
| 480 |
+
"step": 480
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"loss": 0.2023,
|
| 484 |
+
"grad_norm": 0.1960698664188385,
|
| 485 |
+
"learning_rate": 7.691753080453412e-05,
|
| 486 |
+
"entropy": 0.2045104220509529,
|
| 487 |
+
"num_tokens": 7683086.0,
|
| 488 |
+
"mean_token_accuracy": 0.9456699937582016,
|
| 489 |
+
"epoch": 1.6442953020134228,
|
| 490 |
+
"step": 490
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"loss": 0.1991,
|
| 494 |
+
"grad_norm": 0.18973101675510406,
|
| 495 |
+
"learning_rate": 7.60109470525839e-05,
|
| 496 |
+
"entropy": 0.20317464768886567,
|
| 497 |
+
"num_tokens": 7840651.0,
|
| 498 |
+
"mean_token_accuracy": 0.9453895062208175,
|
| 499 |
+
"epoch": 1.6778523489932886,
|
| 500 |
+
"step": 500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"loss": 0.1939,
|
| 504 |
+
"grad_norm": 0.19162558019161224,
|
| 505 |
+
"learning_rate": 7.509248362148889e-05,
|
| 506 |
+
"entropy": 0.19502840861678122,
|
| 507 |
+
"num_tokens": 7997627.0,
|
| 508 |
+
"mean_token_accuracy": 0.9481672704219818,
|
| 509 |
+
"epoch": 1.7114093959731544,
|
| 510 |
+
"step": 510
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"loss": 0.1921,
|
| 514 |
+
"grad_norm": 0.18042920529842377,
|
| 515 |
+
"learning_rate": 7.416255999043401e-05,
|
| 516 |
+
"entropy": 0.19183044098317623,
|
| 517 |
+
"num_tokens": 8153075.0,
|
| 518 |
+
"mean_token_accuracy": 0.9467999368906022,
|
| 519 |
+
"epoch": 1.7449664429530203,
|
| 520 |
+
"step": 520
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"loss": 0.1788,
|
| 524 |
+
"grad_norm": 0.1608305722475052,
|
| 525 |
+
"learning_rate": 7.322160087268877e-05,
|
| 526 |
+
"entropy": 0.18251866959035395,
|
| 527 |
+
"num_tokens": 8307904.0,
|
| 528 |
+
"mean_token_accuracy": 0.9506349295377732,
|
| 529 |
+
"epoch": 1.778523489932886,
|
| 530 |
+
"step": 530
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"loss": 0.1808,
|
| 534 |
+
"grad_norm": 0.18992868065834045,
|
| 535 |
+
"learning_rate": 7.227003602163295e-05,
|
| 536 |
+
"entropy": 0.1834562122821808,
|
| 537 |
+
"num_tokens": 8464236.0,
|
| 538 |
+
"mean_token_accuracy": 0.9505033463239669,
|
| 539 |
+
"epoch": 1.8120805369127517,
|
| 540 |
+
"step": 540
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"loss": 0.1868,
|
| 544 |
+
"grad_norm": 0.21314968168735504,
|
| 545 |
+
"learning_rate": 7.130830003448032e-05,
|
| 546 |
+
"entropy": 0.1886416744440794,
|
| 547 |
+
"num_tokens": 8622442.0,
|
| 548 |
+
"mean_token_accuracy": 0.9492894530296325,
|
| 549 |
+
"epoch": 1.8456375838926173,
|
| 550 |
+
"step": 550
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"loss": 0.19,
|
| 554 |
+
"grad_norm": 0.196988046169281,
|
| 555 |
+
"learning_rate": 7.033683215379002e-05,
|
| 556 |
+
"entropy": 0.1898799568414688,
|
| 557 |
+
"num_tokens": 8778712.0,
|
| 558 |
+
"mean_token_accuracy": 0.9487443953752518,
|
| 559 |
+
"epoch": 1.8791946308724832,
|
| 560 |
+
"step": 560
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"loss": 0.1872,
|
| 564 |
+
"grad_norm": 0.19480957090854645,
|
| 565 |
+
"learning_rate": 6.935607606685642e-05,
|
| 566 |
+
"entropy": 0.1890778660774231,
|
| 567 |
+
"num_tokens": 8934261.0,
|
| 568 |
+
"mean_token_accuracy": 0.9492935538291931,
|
| 569 |
+
"epoch": 1.912751677852349,
|
| 570 |
+
"step": 570
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"loss": 0.1923,
|
| 574 |
+
"grad_norm": 0.1902640014886856,
|
| 575 |
+
"learning_rate": 6.836647970306894e-05,
|
| 576 |
+
"entropy": 0.1931080285459757,
|
| 577 |
+
"num_tokens": 9093573.0,
|
| 578 |
+
"mean_token_accuracy": 0.9487350136041641,
|
| 579 |
+
"epoch": 1.9463087248322148,
|
| 580 |
+
"step": 580
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"loss": 0.1817,
|
| 584 |
+
"grad_norm": 0.19666995108127594,
|
| 585 |
+
"learning_rate": 6.736849502933452e-05,
|
| 586 |
+
"entropy": 0.18354742750525474,
|
| 587 |
+
"num_tokens": 9247665.0,
|
| 588 |
+
"mean_token_accuracy": 0.9502300530672073,
|
| 589 |
+
"epoch": 1.9798657718120807,
|
| 590 |
+
"step": 590
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"loss": 0.1859,
|
| 594 |
+
"grad_norm": 0.2059757262468338,
|
| 595 |
+
"learning_rate": 6.636257784365584e-05,
|
| 596 |
+
"entropy": 0.18999958783388138,
|
| 597 |
+
"num_tokens": 9402433.0,
|
| 598 |
+
"mean_token_accuracy": 0.9492688685655594,
|
| 599 |
+
"epoch": 2.0134228187919465,
|
| 600 |
+
"step": 600
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"loss": 0.1847,
|
| 604 |
+
"grad_norm": 0.1790536344051361,
|
| 605 |
+
"learning_rate": 6.53491875669601e-05,
|
| 606 |
+
"entropy": 0.18768733143806457,
|
| 607 |
+
"num_tokens": 9559144.0,
|
| 608 |
+
"mean_token_accuracy": 0.949212983250618,
|
| 609 |
+
"epoch": 2.046979865771812,
|
| 610 |
+
"step": 610
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"loss": 0.1642,
|
| 614 |
+
"grad_norm": 0.21339593827724457,
|
| 615 |
+
"learning_rate": 6.432878703327298e-05,
|
| 616 |
+
"entropy": 0.16429513394832612,
|
| 617 |
+
"num_tokens": 9714865.0,
|
| 618 |
+
"mean_token_accuracy": 0.9541611701250077,
|
| 619 |
+
"epoch": 2.0805369127516777,
|
| 620 |
+
"step": 620
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"loss": 0.1855,
|
| 624 |
+
"grad_norm": 0.21740688383579254,
|
| 625 |
+
"learning_rate": 6.330184227833376e-05,
|
| 626 |
+
"entropy": 0.18604125529527665,
|
| 627 |
+
"num_tokens": 9872250.0,
|
| 628 |
+
"mean_token_accuracy": 0.9479648023843765,
|
| 629 |
+
"epoch": 2.1140939597315436,
|
| 630 |
+
"step": 630
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"loss": 0.1898,
|
| 634 |
+
"grad_norm": 0.1908544898033142,
|
| 635 |
+
"learning_rate": 6.226882232674825e-05,
|
| 636 |
+
"entropy": 0.19123471677303314,
|
| 637 |
+
"num_tokens": 10026954.0,
|
| 638 |
+
"mean_token_accuracy": 0.9476721823215485,
|
| 639 |
+
"epoch": 2.1476510067114094,
|
| 640 |
+
"step": 640
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"loss": 0.1652,
|
| 644 |
+
"grad_norm": 0.2003369778394699,
|
| 645 |
+
"learning_rate": 6.123019897777657e-05,
|
| 646 |
+
"entropy": 0.16700895316898823,
|
| 647 |
+
"num_tokens": 10177873.0,
|
| 648 |
+
"mean_token_accuracy": 0.9538896352052688,
|
| 649 |
+
"epoch": 2.1812080536912752,
|
| 650 |
+
"step": 650
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"loss": 0.2036,
|
| 654 |
+
"grad_norm": 0.22211240231990814,
|
| 655 |
+
"learning_rate": 6.0186446589853784e-05,
|
| 656 |
+
"entropy": 0.20571780651807786,
|
| 657 |
+
"num_tokens": 10338481.0,
|
| 658 |
+
"mean_token_accuracy": 0.9439575403928757,
|
| 659 |
+
"epoch": 2.214765100671141,
|
| 660 |
+
"step": 660
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"loss": 0.2002,
|
| 664 |
+
"grad_norm": 0.23921112716197968,
|
| 665 |
+
"learning_rate": 5.9138041863941616e-05,
|
| 666 |
+
"entropy": 0.19964271709322928,
|
| 667 |
+
"num_tokens": 10495720.0,
|
| 668 |
+
"mean_token_accuracy": 0.9459815502166748,
|
| 669 |
+
"epoch": 2.248322147651007,
|
| 670 |
+
"step": 670
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"loss": 0.1989,
|
| 674 |
+
"grad_norm": 0.20520520210266113,
|
| 675 |
+
"learning_rate": 5.808546362581032e-05,
|
| 676 |
+
"entropy": 0.20052680410444737,
|
| 677 |
+
"num_tokens": 10655224.0,
|
| 678 |
+
"mean_token_accuracy": 0.9459189057350159,
|
| 679 |
+
"epoch": 2.2818791946308723,
|
| 680 |
+
"step": 680
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"loss": 0.1803,
|
| 684 |
+
"grad_norm": 0.20859530568122864,
|
| 685 |
+
"learning_rate": 5.7029192607350146e-05,
|
| 686 |
+
"entropy": 0.18437107987701892,
|
| 687 |
+
"num_tokens": 10812393.0,
|
| 688 |
+
"mean_token_accuracy": 0.9512098997831344,
|
| 689 |
+
"epoch": 2.315436241610738,
|
| 690 |
+
"step": 690
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"loss": 0.1625,
|
| 694 |
+
"grad_norm": 0.21114759147167206,
|
| 695 |
+
"learning_rate": 5.596971122701221e-05,
|
| 696 |
+
"entropy": 0.16424274519085885,
|
| 697 |
+
"num_tokens": 10965653.0,
|
| 698 |
+
"mean_token_accuracy": 0.9543471187353134,
|
| 699 |
+
"epoch": 2.348993288590604,
|
| 700 |
+
"step": 700
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"loss": 0.164,
|
| 704 |
+
"grad_norm": 0.19510741531848907,
|
| 705 |
+
"learning_rate": 5.4907503369479116e-05,
|
| 706 |
+
"entropy": 0.16679177805781364,
|
| 707 |
+
"num_tokens": 11120539.0,
|
| 708 |
+
"mean_token_accuracy": 0.954514691233635,
|
| 709 |
+
"epoch": 2.38255033557047,
|
| 710 |
+
"step": 710
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"loss": 0.1772,
|
| 714 |
+
"grad_norm": 0.244350403547287,
|
| 715 |
+
"learning_rate": 5.384305416466584e-05,
|
| 716 |
+
"entropy": 0.17797317542135715,
|
| 717 |
+
"num_tokens": 11275790.0,
|
| 718 |
+
"mean_token_accuracy": 0.950087907910347,
|
| 719 |
+
"epoch": 2.4161073825503356,
|
| 720 |
+
"step": 720
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"loss": 0.1835,
|
| 724 |
+
"grad_norm": 0.19845031201839447,
|
| 725 |
+
"learning_rate": 5.2776849766152e-05,
|
| 726 |
+
"entropy": 0.18638909384608268,
|
| 727 |
+
"num_tokens": 11430204.0,
|
| 728 |
+
"mean_token_accuracy": 0.9490520358085632,
|
| 729 |
+
"epoch": 2.4496644295302015,
|
| 730 |
+
"step": 730
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"loss": 0.1883,
|
| 734 |
+
"grad_norm": 0.20039933919906616,
|
| 735 |
+
"learning_rate": 5.170937712914655e-05,
|
| 736 |
+
"entropy": 0.18770763501524926,
|
| 737 |
+
"num_tokens": 11592345.0,
|
| 738 |
+
"mean_token_accuracy": 0.9483928889036178,
|
| 739 |
+
"epoch": 2.4832214765100673,
|
| 740 |
+
"step": 740
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"loss": 0.1528,
|
| 744 |
+
"grad_norm": 0.20752805471420288,
|
| 745 |
+
"learning_rate": 5.064112378808637e-05,
|
| 746 |
+
"entropy": 0.15445428304374217,
|
| 747 |
+
"num_tokens": 11744570.0,
|
| 748 |
+
"mean_token_accuracy": 0.9564135998487473,
|
| 749 |
+
"epoch": 2.5167785234899327,
|
| 750 |
+
"step": 750
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"loss": 0.1793,
|
| 754 |
+
"grad_norm": 0.2342364341020584,
|
| 755 |
+
"learning_rate": 4.957257763397024e-05,
|
| 756 |
+
"entropy": 0.18113909810781478,
|
| 757 |
+
"num_tokens": 11901062.0,
|
| 758 |
+
"mean_token_accuracy": 0.9501085758209229,
|
| 759 |
+
"epoch": 2.5503355704697985,
|
| 760 |
+
"step": 760
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"loss": 0.1572,
|
| 764 |
+
"grad_norm": 0.20083309710025787,
|
| 765 |
+
"learning_rate": 4.850422669153009e-05,
|
| 766 |
+
"entropy": 0.16168876476585864,
|
| 767 |
+
"num_tokens": 12053753.0,
|
| 768 |
+
"mean_token_accuracy": 0.9557742238044739,
|
| 769 |
+
"epoch": 2.5838926174496644,
|
| 770 |
+
"step": 770
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"loss": 0.1625,
|
| 774 |
+
"grad_norm": 0.20891757309436798,
|
| 775 |
+
"learning_rate": 4.743655889634105e-05,
|
| 776 |
+
"entropy": 0.16362526044249534,
|
| 777 |
+
"num_tokens": 12208995.0,
|
| 778 |
+
"mean_token_accuracy": 0.9552077889442444,
|
| 779 |
+
"epoch": 2.61744966442953,
|
| 780 |
+
"step": 780
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"loss": 0.1567,
|
| 784 |
+
"grad_norm": 0.21302001178264618,
|
| 785 |
+
"learning_rate": 4.6370061871972326e-05,
|
| 786 |
+
"entropy": 0.15993254519999028,
|
| 787 |
+
"num_tokens": 12364530.0,
|
| 788 |
+
"mean_token_accuracy": 0.955506107211113,
|
| 789 |
+
"epoch": 2.651006711409396,
|
| 790 |
+
"step": 790
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"loss": 0.1705,
|
| 794 |
+
"grad_norm": 0.19647525250911713,
|
| 795 |
+
"learning_rate": 4.530522270728048e-05,
|
| 796 |
+
"entropy": 0.1731419090181589,
|
| 797 |
+
"num_tokens": 12522512.0,
|
| 798 |
+
"mean_token_accuracy": 0.9523056238889694,
|
| 799 |
+
"epoch": 2.684563758389262,
|
| 800 |
+
"step": 800
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"loss": 0.1656,
|
| 804 |
+
"grad_norm": 0.23564544320106506,
|
| 805 |
+
"learning_rate": 4.424252773394704e-05,
|
| 806 |
+
"entropy": 0.16778925359249114,
|
| 807 |
+
"num_tokens": 12678022.0,
|
| 808 |
+
"mean_token_accuracy": 0.9539547711610794,
|
| 809 |
+
"epoch": 2.7181208053691277,
|
| 810 |
+
"step": 810
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"loss": 0.1862,
|
| 814 |
+
"grad_norm": 0.19549453258514404,
|
| 815 |
+
"learning_rate": 4.318246230436174e-05,
|
| 816 |
+
"entropy": 0.1878517519682646,
|
| 817 |
+
"num_tokens": 12838063.0,
|
| 818 |
+
"mean_token_accuracy": 0.949842843413353,
|
| 819 |
+
"epoch": 2.751677852348993,
|
| 820 |
+
"step": 820
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"loss": 0.1687,
|
| 824 |
+
"grad_norm": 0.23449555039405823,
|
| 825 |
+
"learning_rate": 4.212551056995323e-05,
|
| 826 |
+
"entropy": 0.17150973305106162,
|
| 827 |
+
"num_tokens": 12993073.0,
|
| 828 |
+
"mean_token_accuracy": 0.953241491317749,
|
| 829 |
+
"epoch": 2.785234899328859,
|
| 830 |
+
"step": 830
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"loss": 0.181,
|
| 834 |
+
"grad_norm": 0.19433574378490448,
|
| 835 |
+
"learning_rate": 4.107215526006817e-05,
|
| 836 |
+
"entropy": 0.1791513752192259,
|
| 837 |
+
"num_tokens": 13151000.0,
|
| 838 |
+
"mean_token_accuracy": 0.9498830765485764,
|
| 839 |
+
"epoch": 2.8187919463087248,
|
| 840 |
+
"step": 840
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"loss": 0.1782,
|
| 844 |
+
"grad_norm": 0.31362298130989075,
|
| 845 |
+
"learning_rate": 4.00228774614998e-05,
|
| 846 |
+
"entropy": 0.18104747980833052,
|
| 847 |
+
"num_tokens": 13306740.0,
|
| 848 |
+
"mean_token_accuracy": 0.9511455327272416,
|
| 849 |
+
"epoch": 2.8523489932885906,
|
| 850 |
+
"step": 850
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"loss": 0.1841,
|
| 854 |
+
"grad_norm": 0.19440777599811554,
|
| 855 |
+
"learning_rate": 3.897815639876673e-05,
|
| 856 |
+
"entropy": 0.183981429412961,
|
| 857 |
+
"num_tokens": 13465413.0,
|
| 858 |
+
"mean_token_accuracy": 0.9499198466539382,
|
| 859 |
+
"epoch": 2.8859060402684564,
|
| 860 |
+
"step": 860
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"loss": 0.1727,
|
| 864 |
+
"grad_norm": 0.22845511138439178,
|
| 865 |
+
"learning_rate": 3.793846921524237e-05,
|
| 866 |
+
"entropy": 0.17417334467172624,
|
| 867 |
+
"num_tokens": 13625300.0,
|
| 868 |
+
"mean_token_accuracy": 0.9522460520267486,
|
| 869 |
+
"epoch": 2.9194630872483223,
|
| 870 |
+
"step": 870
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"loss": 0.1917,
|
| 874 |
+
"grad_norm": 0.17185327410697937,
|
| 875 |
+
"learning_rate": 3.6904290755234604e-05,
|
| 876 |
+
"entropy": 0.19201257564127444,
|
| 877 |
+
"num_tokens": 13787633.0,
|
| 878 |
+
"mean_token_accuracy": 0.9470277667045593,
|
| 879 |
+
"epoch": 2.953020134228188,
|
| 880 |
+
"step": 880
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"loss": 0.181,
|
| 884 |
+
"grad_norm": 0.22177894413471222,
|
| 885 |
+
"learning_rate": 3.587609334711576e-05,
|
| 886 |
+
"entropy": 0.18371602855622768,
|
| 887 |
+
"num_tokens": 13946986.0,
|
| 888 |
+
"mean_token_accuracy": 0.9502452671527862,
|
| 889 |
+
"epoch": 2.9865771812080535,
|
| 890 |
+
"step": 890
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"loss": 0.1639,
|
| 894 |
+
"grad_norm": 0.20057617127895355,
|
| 895 |
+
"learning_rate": 3.48543465876014e-05,
|
| 896 |
+
"entropy": 0.16733385995030403,
|
| 897 |
+
"num_tokens": 14102933.0,
|
| 898 |
+
"mean_token_accuracy": 0.9548174262046814,
|
| 899 |
+
"epoch": 3.0201342281879193,
|
| 900 |
+
"step": 900
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"loss": 0.1657,
|
| 904 |
+
"grad_norm": 0.234299898147583,
|
| 905 |
+
"learning_rate": 3.383951712727701e-05,
|
| 906 |
+
"entropy": 0.1675445105880499,
|
| 907 |
+
"num_tokens": 14259753.0,
|
| 908 |
+
"mean_token_accuracy": 0.9535206586122513,
|
| 909 |
+
"epoch": 3.053691275167785,
|
| 910 |
+
"step": 910
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"loss": 0.1583,
|
| 914 |
+
"grad_norm": 0.23432078957557678,
|
| 915 |
+
"learning_rate": 3.2832068457469945e-05,
|
| 916 |
+
"entropy": 0.1605758022516966,
|
| 917 |
+
"num_tokens": 14418531.0,
|
| 918 |
+
"mean_token_accuracy": 0.955757224559784,
|
| 919 |
+
"epoch": 3.087248322147651,
|
| 920 |
+
"step": 920
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"loss": 0.1504,
|
| 924 |
+
"grad_norm": 0.20643459260463715,
|
| 925 |
+
"learning_rate": 3.183246069856443e-05,
|
| 926 |
+
"entropy": 0.15479738190770148,
|
| 927 |
+
"num_tokens": 14573081.0,
|
| 928 |
+
"mean_token_accuracy": 0.9577652394771576,
|
| 929 |
+
"epoch": 3.120805369127517,
|
| 930 |
+
"step": 930
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"loss": 0.1616,
|
| 934 |
+
"grad_norm": 0.2060076892375946,
|
| 935 |
+
"learning_rate": 3.0841150389856125e-05,
|
| 936 |
+
"entropy": 0.16488183476030827,
|
| 937 |
+
"num_tokens": 14727805.0,
|
| 938 |
+
"mean_token_accuracy": 0.9542657196521759,
|
| 939 |
+
"epoch": 3.1543624161073827,
|
| 940 |
+
"step": 940
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"loss": 0.1727,
|
| 944 |
+
"grad_norm": 0.25925537943840027,
|
| 945 |
+
"learning_rate": 2.9858590281042152e-05,
|
| 946 |
+
"entropy": 0.17419612146914004,
|
| 947 |
+
"num_tokens": 14883574.0,
|
| 948 |
+
"mean_token_accuracy": 0.9522079944610595,
|
| 949 |
+
"epoch": 3.1879194630872485,
|
| 950 |
+
"step": 950
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"loss": 0.1444,
|
| 954 |
+
"grad_norm": 0.22550174593925476,
|
| 955 |
+
"learning_rate": 2.888522912544202e-05,
|
| 956 |
+
"entropy": 0.1470055654644966,
|
| 957 |
+
"num_tokens": 15038732.0,
|
| 958 |
+
"mean_token_accuracy": 0.9591749131679534,
|
| 959 |
+
"epoch": 3.221476510067114,
|
| 960 |
+
"step": 960
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"loss": 0.1453,
|
| 964 |
+
"grad_norm": 0.22862796485424042,
|
| 965 |
+
"learning_rate": 2.792151147504366e-05,
|
| 966 |
+
"entropy": 0.14832949005067347,
|
| 967 |
+
"num_tokens": 15193067.0,
|
| 968 |
+
"mean_token_accuracy": 0.9592784523963929,
|
| 969 |
+
"epoch": 3.2550335570469797,
|
| 970 |
+
"step": 970
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"loss": 0.1592,
|
| 974 |
+
"grad_norm": 0.21343277394771576,
|
| 975 |
+
"learning_rate": 2.6967877477468397e-05,
|
| 976 |
+
"entropy": 0.16025195196270942,
|
| 977 |
+
"num_tokens": 15348668.0,
|
| 978 |
+
"mean_token_accuracy": 0.9546813875436783,
|
| 979 |
+
"epoch": 3.2885906040268456,
|
| 980 |
+
"step": 980
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"loss": 0.1842,
|
| 984 |
+
"grad_norm": 0.27152112126350403,
|
| 985 |
+
"learning_rate": 2.6024762674947313e-05,
|
| 986 |
+
"entropy": 0.18718509674072265,
|
| 987 |
+
"num_tokens": 15507728.0,
|
| 988 |
+
"mean_token_accuracy": 0.9490089803934098,
|
| 989 |
+
"epoch": 3.3221476510067114,
|
| 990 |
+
"step": 990
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"loss": 0.1613,
|
| 994 |
+
"grad_norm": 0.20685099065303802,
|
| 995 |
+
"learning_rate": 2.509259780540118e-05,
|
| 996 |
+
"entropy": 0.16506212055683137,
|
| 997 |
+
"num_tokens": 15663068.0,
|
| 998 |
+
"mean_token_accuracy": 0.9536582648754119,
|
| 999 |
+
"epoch": 3.3557046979865772,
|
| 1000 |
+
"step": 1000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"loss": 0.1557,
|
| 1004 |
+
"grad_norm": 0.24159930646419525,
|
| 1005 |
+
"learning_rate": 2.4171808605714504e-05,
|
| 1006 |
+
"entropy": 0.15720976293087005,
|
| 1007 |
+
"num_tokens": 15820651.0,
|
| 1008 |
+
"mean_token_accuracy": 0.9559206962585449,
|
| 1009 |
+
"epoch": 3.389261744966443,
|
| 1010 |
+
"step": 1010
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"loss": 0.1775,
|
| 1014 |
+
"grad_norm": 0.23764103651046753,
|
| 1015 |
+
"learning_rate": 2.3262815617293517e-05,
|
| 1016 |
+
"entropy": 0.1807922873646021,
|
| 1017 |
+
"num_tokens": 15979358.0,
|
| 1018 |
+
"mean_token_accuracy": 0.9511760205030442,
|
| 1019 |
+
"epoch": 3.422818791946309,
|
| 1020 |
+
"step": 1020
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"loss": 0.1713,
|
| 1024 |
+
"grad_norm": 0.22108638286590576,
|
| 1025 |
+
"learning_rate": 2.2366033993997344e-05,
|
| 1026 |
+
"entropy": 0.17095785699784755,
|
| 1027 |
+
"num_tokens": 16135558.0,
|
| 1028 |
+
"mean_token_accuracy": 0.95192691385746,
|
| 1029 |
+
"epoch": 3.4563758389261743,
|
| 1030 |
+
"step": 1030
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"loss": 0.1682,
|
| 1034 |
+
"grad_norm": 0.24025718867778778,
|
| 1035 |
+
"learning_rate": 2.1481873312529426e-05,
|
| 1036 |
+
"entropy": 0.17155840508639814,
|
| 1037 |
+
"num_tokens": 16289493.0,
|
| 1038 |
+
"mean_token_accuracy": 0.9537844061851501,
|
| 1039 |
+
"epoch": 3.48993288590604,
|
| 1040 |
+
"step": 1040
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"loss": 0.1645,
|
| 1044 |
+
"grad_norm": 0.22691352665424347,
|
| 1045 |
+
"learning_rate": 2.061073738537635e-05,
|
| 1046 |
+
"entropy": 0.1673688843846321,
|
| 1047 |
+
"num_tokens": 16444253.0,
|
| 1048 |
+
"mean_token_accuracy": 0.9540765851736068,
|
| 1049 |
+
"epoch": 3.523489932885906,
|
| 1050 |
+
"step": 1050
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"loss": 0.1711,
|
| 1054 |
+
"grad_norm": 0.260170042514801,
|
| 1055 |
+
"learning_rate": 1.975302407637929e-05,
|
| 1056 |
+
"entropy": 0.17190844528377056,
|
| 1057 |
+
"num_tokens": 16605899.0,
|
| 1058 |
+
"mean_token_accuracy": 0.9533313482999801,
|
| 1059 |
+
"epoch": 3.557046979865772,
|
| 1060 |
+
"step": 1060
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"loss": 0.1703,
|
| 1064 |
+
"grad_norm": 0.22882592678070068,
|
| 1065 |
+
"learning_rate": 1.8909125119022136e-05,
|
| 1066 |
+
"entropy": 0.17202796116471292,
|
| 1067 |
+
"num_tokens": 16762980.0,
|
| 1068 |
+
"mean_token_accuracy": 0.9525971114635468,
|
| 1069 |
+
"epoch": 3.5906040268456376,
|
| 1070 |
+
"step": 1070
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"loss": 0.1751,
|
| 1074 |
+
"grad_norm": 0.2285274714231491,
|
| 1075 |
+
"learning_rate": 1.807942593751973e-05,
|
| 1076 |
+
"entropy": 0.17774682715535164,
|
| 1077 |
+
"num_tokens": 16923345.0,
|
| 1078 |
+
"mean_token_accuracy": 0.9512902319431304,
|
| 1079 |
+
"epoch": 3.6241610738255035,
|
| 1080 |
+
"step": 1080
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"loss": 0.1577,
|
| 1084 |
+
"grad_norm": 0.24551482498645782,
|
| 1085 |
+
"learning_rate": 1.7264305470787363e-05,
|
| 1086 |
+
"entropy": 0.1609206810593605,
|
| 1087 |
+
"num_tokens": 17077331.0,
|
| 1088 |
+
"mean_token_accuracy": 0.955593678355217,
|
| 1089 |
+
"epoch": 3.6577181208053693,
|
| 1090 |
+
"step": 1090
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"loss": 0.1558,
|
| 1094 |
+
"grad_norm": 0.22661259770393372,
|
| 1095 |
+
"learning_rate": 1.6464135999372537e-05,
|
| 1096 |
+
"entropy": 0.1576618704944849,
|
| 1097 |
+
"num_tokens": 17231136.0,
|
| 1098 |
+
"mean_token_accuracy": 0.9559727132320404,
|
| 1099 |
+
"epoch": 3.6912751677852347,
|
| 1100 |
+
"step": 1100
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"loss": 0.187,
|
| 1104 |
+
"grad_norm": 0.23124344646930695,
|
| 1105 |
+
"learning_rate": 1.567928297542749e-05,
|
| 1106 |
+
"entropy": 0.18841546326875686,
|
| 1107 |
+
"num_tokens": 17390871.0,
|
| 1108 |
+
"mean_token_accuracy": 0.9487693756818771,
|
| 1109 |
+
"epoch": 3.7248322147651005,
|
| 1110 |
+
"step": 1110
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"loss": 0.1368,
|
| 1114 |
+
"grad_norm": 0.2734222710132599,
|
| 1115 |
+
"learning_rate": 1.4910104855800427e-05,
|
| 1116 |
+
"entropy": 0.14009847678244114,
|
| 1117 |
+
"num_tokens": 17544815.0,
|
| 1118 |
+
"mean_token_accuracy": 0.9597841829061509,
|
| 1119 |
+
"epoch": 3.7583892617449663,
|
| 1120 |
+
"step": 1120
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"loss": 0.1557,
|
| 1124 |
+
"grad_norm": 0.25066235661506653,
|
| 1125 |
+
"learning_rate": 1.4156952938321798e-05,
|
| 1126 |
+
"entropy": 0.1570753049105406,
|
| 1127 |
+
"num_tokens": 17701337.0,
|
| 1128 |
+
"mean_token_accuracy": 0.9568644523620605,
|
| 1129 |
+
"epoch": 3.791946308724832,
|
| 1130 |
+
"step": 1130
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"loss": 0.1582,
|
| 1134 |
+
"grad_norm": 0.24784211814403534,
|
| 1135 |
+
"learning_rate": 1.3420171201359933e-05,
|
| 1136 |
+
"entropy": 0.15978324972093105,
|
| 1137 |
+
"num_tokens": 17855471.0,
|
| 1138 |
+
"mean_token_accuracy": 0.9565310895442962,
|
| 1139 |
+
"epoch": 3.825503355704698,
|
| 1140 |
+
"step": 1140
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"loss": 0.1583,
|
| 1144 |
+
"grad_norm": 0.22498124837875366,
|
| 1145 |
+
"learning_rate": 1.2700096146719931e-05,
|
| 1146 |
+
"entropy": 0.16122703738510608,
|
| 1147 |
+
"num_tokens": 18017269.0,
|
| 1148 |
+
"mean_token_accuracy": 0.955768808722496,
|
| 1149 |
+
"epoch": 3.859060402684564,
|
| 1150 |
+
"step": 1150
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"loss": 0.1729,
|
| 1154 |
+
"grad_norm": 0.22251562774181366,
|
| 1155 |
+
"learning_rate": 1.1997056645956967e-05,
|
| 1156 |
+
"entropy": 0.17401745431125165,
|
| 1157 |
+
"num_tokens": 18176540.0,
|
| 1158 |
+
"mean_token_accuracy": 0.9513998061418534,
|
| 1159 |
+
"epoch": 3.8926174496644297,
|
| 1160 |
+
"step": 1160
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"loss": 0.1429,
|
| 1164 |
+
"grad_norm": 0.2645820677280426,
|
| 1165 |
+
"learning_rate": 1.1311373790174657e-05,
|
| 1166 |
+
"entropy": 0.14712858721613883,
|
| 1167 |
+
"num_tokens": 18331179.0,
|
| 1168 |
+
"mean_token_accuracy": 0.9599408686161042,
|
| 1169 |
+
"epoch": 3.926174496644295,
|
| 1170 |
+
"step": 1170
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"loss": 0.1778,
|
| 1174 |
+
"grad_norm": 0.22513671219348907,
|
| 1175 |
+
"learning_rate": 1.0643360743376829e-05,
|
| 1176 |
+
"entropy": 0.178417731449008,
|
| 1177 |
+
"num_tokens": 18491744.0,
|
| 1178 |
+
"mean_token_accuracy": 0.9512795448303223,
|
| 1179 |
+
"epoch": 3.959731543624161,
|
| 1180 |
+
"step": 1180
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"loss": 0.1751,
|
| 1184 |
+
"grad_norm": 0.23683220148086548,
|
| 1185 |
+
"learning_rate": 9.993322599439692e-06,
|
| 1186 |
+
"entropy": 0.1774978566914797,
|
| 1187 |
+
"num_tokens": 18647922.0,
|
| 1188 |
+
"mean_token_accuracy": 0.9514397591352463,
|
| 1189 |
+
"epoch": 3.9932885906040267,
|
| 1190 |
+
"step": 1190
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"loss": 0.1343,
|
| 1194 |
+
"grad_norm": 0.23946121335029602,
|
| 1195 |
+
"learning_rate": 9.36155624276987e-06,
|
| 1196 |
+
"entropy": 0.13949108868837357,
|
| 1197 |
+
"num_tokens": 18799741.0,
|
| 1198 |
+
"mean_token_accuracy": 0.961316853761673,
|
| 1199 |
+
"epoch": 4.026845637583893,
|
| 1200 |
+
"step": 1200
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"loss": 0.1448,
|
| 1204 |
+
"grad_norm": 0.26739007234573364,
|
| 1205 |
+
"learning_rate": 8.748350212711853e-06,
|
| 1206 |
+
"entropy": 0.14928585700690747,
|
| 1207 |
+
"num_tokens": 18954239.0,
|
| 1208 |
+
"mean_token_accuracy": 0.9586472451686859,
|
| 1209 |
+
"epoch": 4.060402684563758,
|
| 1210 |
+
"step": 1210
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"loss": 0.1589,
|
| 1214 |
+
"grad_norm": 0.2560782730579376,
|
| 1215 |
+
"learning_rate": 8.153984571766699e-06,
|
| 1216 |
+
"entropy": 0.1610208224505186,
|
| 1217 |
+
"num_tokens": 19113299.0,
|
| 1218 |
+
"mean_token_accuracy": 0.955417987704277,
|
| 1219 |
+
"epoch": 4.093959731543624,
|
| 1220 |
+
"step": 1220
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"loss": 0.1566,
|
| 1224 |
+
"grad_norm": 0.2503839135169983,
|
| 1225 |
+
"learning_rate": 7.578730777682386e-06,
|
| 1226 |
+
"entropy": 0.15893662311136722,
|
| 1227 |
+
"num_tokens": 19269704.0,
|
| 1228 |
+
"mean_token_accuracy": 0.9564489126205444,
|
| 1229 |
+
"epoch": 4.12751677852349,
|
| 1230 |
+
"step": 1230
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"loss": 0.1729,
|
| 1234 |
+
"grad_norm": 0.2269158661365509,
|
| 1235 |
+
"learning_rate": 7.022851559473964e-06,
|
| 1236 |
+
"entropy": 0.17561149559915065,
|
| 1237 |
+
"num_tokens": 19430349.0,
|
| 1238 |
+
"mean_token_accuracy": 0.95240318775177,
|
| 1239 |
+
"epoch": 4.1610738255033555,
|
| 1240 |
+
"step": 1240
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"loss": 0.168,
|
| 1244 |
+
"grad_norm": 0.2869897186756134,
|
| 1245 |
+
"learning_rate": 6.486600797430492e-06,
|
| 1246 |
+
"entropy": 0.1709251943975687,
|
| 1247 |
+
"num_tokens": 19585660.0,
|
| 1248 |
+
"mean_token_accuracy": 0.9534720987081527,
|
| 1249 |
+
"epoch": 4.194630872483222,
|
| 1250 |
+
"step": 1250
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"loss": 0.1457,
|
| 1254 |
+
"grad_norm": 0.25000014901161194,
|
| 1255 |
+
"learning_rate": 5.9702234071631e-06,
|
| 1256 |
+
"entropy": 0.14829469434916973,
|
| 1257 |
+
"num_tokens": 19744182.0,
|
| 1258 |
+
"mean_token_accuracy": 0.9593781590461731,
|
| 1259 |
+
"epoch": 4.228187919463087,
|
| 1260 |
+
"step": 1260
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"loss": 0.16,
|
| 1264 |
+
"grad_norm": 0.24967129528522491,
|
| 1265 |
+
"learning_rate": 5.473955227747623e-06,
|
| 1266 |
+
"entropy": 0.16518744193017482,
|
| 1267 |
+
"num_tokens": 19902824.0,
|
| 1268 |
+
"mean_token_accuracy": 0.9548085004091262,
|
| 1269 |
+
"epoch": 4.261744966442953,
|
| 1270 |
+
"step": 1270
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"loss": 0.1508,
|
| 1274 |
+
"grad_norm": 0.27587053179740906,
|
| 1275 |
+
"learning_rate": 4.9980229140124905e-06,
|
| 1276 |
+
"entropy": 0.15380522198975086,
|
| 1277 |
+
"num_tokens": 20057944.0,
|
| 1278 |
+
"mean_token_accuracy": 0.9581666350364685,
|
| 1279 |
+
"epoch": 4.295302013422819,
|
| 1280 |
+
"step": 1280
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"loss": 0.1497,
|
| 1284 |
+
"grad_norm": 0.2772783935070038,
|
| 1285 |
+
"learning_rate": 4.542643833021254e-06,
|
| 1286 |
+
"entropy": 0.15263673029839991,
|
| 1287 |
+
"num_tokens": 20215072.0,
|
| 1288 |
+
"mean_token_accuracy": 0.9581062227487565,
|
| 1289 |
+
"epoch": 4.328859060402684,
|
| 1290 |
+
"step": 1290
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"loss": 0.1313,
|
| 1294 |
+
"grad_norm": 0.25381705164909363,
|
| 1295 |
+
"learning_rate": 4.108025964797135e-06,
|
| 1296 |
+
"entropy": 0.13392407521605493,
|
| 1297 |
+
"num_tokens": 20371415.0,
|
| 1298 |
+
"mean_token_accuracy": 0.9618801653385163,
|
| 1299 |
+
"epoch": 4.3624161073825505,
|
| 1300 |
+
"step": 1300
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"loss": 0.1442,
|
| 1304 |
+
"grad_norm": 0.318499892950058,
|
| 1305 |
+
"learning_rate": 3.69436780733462e-06,
|
| 1306 |
+
"entropy": 0.1467492014169693,
|
| 1307 |
+
"num_tokens": 20525796.0,
|
| 1308 |
+
"mean_token_accuracy": 0.9589717149734497,
|
| 1309 |
+
"epoch": 4.395973154362416,
|
| 1310 |
+
"step": 1310
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"loss": 0.164,
|
| 1314 |
+
"grad_norm": 0.29794761538505554,
|
| 1315 |
+
"learning_rate": 3.3018582859418446e-06,
|
| 1316 |
+
"entropy": 0.16693559624254703,
|
| 1317 |
+
"num_tokens": 20683012.0,
|
| 1318 |
+
"mean_token_accuracy": 0.9544567495584488,
|
| 1319 |
+
"epoch": 4.429530201342282,
|
| 1320 |
+
"step": 1320
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"loss": 0.1585,
|
| 1324 |
+
"grad_norm": 0.26992982625961304,
|
| 1325 |
+
"learning_rate": 2.930676666954846e-06,
|
| 1326 |
+
"entropy": 0.16096150763332845,
|
| 1327 |
+
"num_tokens": 20836880.0,
|
| 1328 |
+
"mean_token_accuracy": 0.955687940120697,
|
| 1329 |
+
"epoch": 4.4630872483221475,
|
| 1330 |
+
"step": 1330
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"loss": 0.1545,
|
| 1334 |
+
"grad_norm": 0.2633298635482788,
|
| 1335 |
+
"learning_rate": 2.580992475863381e-06,
|
| 1336 |
+
"entropy": 0.15808030292391778,
|
| 1337 |
+
"num_tokens": 20991520.0,
|
| 1338 |
+
"mean_token_accuracy": 0.9564832031726838,
|
| 1339 |
+
"epoch": 4.496644295302014,
|
| 1340 |
+
"step": 1340
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"loss": 0.1574,
|
| 1344 |
+
"grad_norm": 0.2697800099849701,
|
| 1345 |
+
"learning_rate": 2.2529654198854835e-06,
|
| 1346 |
+
"entropy": 0.16020409800112248,
|
| 1347 |
+
"num_tokens": 21149049.0,
|
| 1348 |
+
"mean_token_accuracy": 0.9553818762302398,
|
| 1349 |
+
"epoch": 4.530201342281879,
|
| 1350 |
+
"step": 1350
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"loss": 0.1434,
|
| 1354 |
+
"grad_norm": 0.26713675260543823,
|
| 1355 |
+
"learning_rate": 1.9467453150262327e-06,
|
| 1356 |
+
"entropy": 0.1468834660947323,
|
| 1357 |
+
"num_tokens": 21303482.0,
|
| 1358 |
+
"mean_token_accuracy": 0.9584787577390671,
|
| 1359 |
+
"epoch": 4.563758389261745,
|
| 1360 |
+
"step": 1360
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"loss": 0.1676,
|
| 1364 |
+
"grad_norm": 0.25141027569770813,
|
| 1365 |
+
"learning_rate": 1.6624720176540265e-06,
|
| 1366 |
+
"entropy": 0.17059922069311143,
|
| 1367 |
+
"num_tokens": 21462214.0,
|
| 1368 |
+
"mean_token_accuracy": 0.9529773116111755,
|
| 1369 |
+
"epoch": 4.597315436241611,
|
| 1370 |
+
"step": 1370
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"loss": 0.1631,
|
| 1374 |
+
"grad_norm": 0.22781124711036682,
|
| 1375 |
+
"learning_rate": 1.400275360625608e-06,
|
| 1376 |
+
"entropy": 0.16469677537679672,
|
| 1377 |
+
"num_tokens": 21617768.0,
|
| 1378 |
+
"mean_token_accuracy": 0.9540845274925231,
|
| 1379 |
+
"epoch": 4.630872483221476,
|
| 1380 |
+
"step": 1380
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"loss": 0.1405,
|
| 1384 |
+
"grad_norm": 0.26300156116485596,
|
| 1385 |
+
"learning_rate": 1.1602750939889774e-06,
|
| 1386 |
+
"entropy": 0.14349478296935558,
|
| 1387 |
+
"num_tokens": 21775827.0,
|
| 1388 |
+
"mean_token_accuracy": 0.9608874082565307,
|
| 1389 |
+
"epoch": 4.6644295302013425,
|
| 1390 |
+
"step": 1390
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"loss": 0.1648,
|
| 1394 |
+
"grad_norm": 0.28018462657928467,
|
| 1395 |
+
"learning_rate": 9.42580830291373e-07,
|
| 1396 |
+
"entropy": 0.168378734216094,
|
| 1397 |
+
"num_tokens": 21933598.0,
|
| 1398 |
+
"mean_token_accuracy": 0.9548741400241851,
|
| 1399 |
+
"epoch": 4.697986577181208,
|
| 1400 |
+
"step": 1400
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"loss": 0.1579,
|
| 1404 |
+
"grad_norm": 0.23573631048202515,
|
| 1405 |
+
"learning_rate": 7.472919945171631e-07,
|
| 1406 |
+
"entropy": 0.16024879328906536,
|
| 1407 |
+
"num_tokens": 22092356.0,
|
| 1408 |
+
"mean_token_accuracy": 0.9563170611858368,
|
| 1409 |
+
"epoch": 4.731543624161074,
|
| 1410 |
+
"step": 1410
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"loss": 0.1486,
|
| 1414 |
+
"grad_norm": 0.26423606276512146,
|
| 1415 |
+
"learning_rate": 5.74497778678662e-07,
|
| 1416 |
+
"entropy": 0.153883695602417,
|
| 1417 |
+
"num_tokens": 22245245.0,
|
| 1418 |
+
"mean_token_accuracy": 0.9581821829080581,
|
| 1419 |
+
"epoch": 4.76510067114094,
|
| 1420 |
+
"step": 1420
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"loss": 0.1668,
|
| 1424 |
+
"grad_norm": 0.24437865614891052,
|
| 1425 |
+
"learning_rate": 4.242771010804558e-07,
|
| 1426 |
+
"entropy": 0.16868923045694828,
|
| 1427 |
+
"num_tokens": 22403545.0,
|
| 1428 |
+
"mean_token_accuracy": 0.9531532049179077,
|
| 1429 |
+
"epoch": 4.798657718120805,
|
| 1430 |
+
"step": 1430
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"loss": 0.1497,
|
| 1434 |
+
"grad_norm": 0.2679818868637085,
|
| 1435 |
+
"learning_rate": 2.966985702759828e-07,
|
| 1436 |
+
"entropy": 0.15332721956074238,
|
| 1437 |
+
"num_tokens": 22559658.0,
|
| 1438 |
+
"mean_token_accuracy": 0.9572977095842361,
|
| 1439 |
+
"epoch": 4.832214765100671,
|
| 1440 |
+
"step": 1440
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"loss": 0.1487,
|
| 1444 |
+
"grad_norm": 0.25875014066696167,
|
| 1445 |
+
"learning_rate": 1.9182045373273838e-07,
|
| 1446 |
+
"entropy": 0.15049569718539715,
|
| 1447 |
+
"num_tokens": 22718226.0,
|
| 1448 |
+
"mean_token_accuracy": 0.9580355733633041,
|
| 1449 |
+
"epoch": 4.865771812080537,
|
| 1450 |
+
"step": 1450
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"loss": 0.1444,
|
| 1454 |
+
"grad_norm": 0.25537535548210144,
|
| 1455 |
+
"learning_rate": 1.0969065122041766e-07,
|
| 1456 |
+
"entropy": 0.14743491858243943,
|
| 1457 |
+
"num_tokens": 22873841.0,
|
| 1458 |
+
"mean_token_accuracy": 0.9597749501466751,
|
| 1459 |
+
"epoch": 4.899328859060403,
|
| 1460 |
+
"step": 1460
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"loss": 0.1584,
|
| 1464 |
+
"grad_norm": 0.28133252263069153,
|
| 1465 |
+
"learning_rate": 5.0346672934270534e-08,
|
| 1466 |
+
"entropy": 0.16207893192768097,
|
| 1467 |
+
"num_tokens": 23032944.0,
|
| 1468 |
+
"mean_token_accuracy": 0.9557940989732743,
|
| 1469 |
+
"epoch": 4.932885906040268,
|
| 1470 |
+
"step": 1470
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"loss": 0.1577,
|
| 1474 |
+
"grad_norm": 0.25087353587150574,
|
| 1475 |
+
"learning_rate": 1.3815622363427815e-08,
|
| 1476 |
+
"entropy": 0.16037070676684378,
|
| 1477 |
+
"num_tokens": 23191641.0,
|
| 1478 |
+
"mean_token_accuracy": 0.9560932219028473,
|
| 1479 |
+
"epoch": 4.966442953020135,
|
| 1480 |
+
"step": 1480
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"loss": 0.1779,
|
| 1484 |
+
"grad_norm": 0.307130366563797,
|
| 1485 |
+
"learning_rate": 1.141839123142141e-10,
|
| 1486 |
+
"entropy": 0.18243111334741116,
|
| 1487 |
+
"num_tokens": 23348865.0,
|
| 1488 |
+
"mean_token_accuracy": 0.9507445961236953,
|
| 1489 |
+
"epoch": 5.0,
|
| 1490 |
+
"step": 1490
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"train_runtime": 10055.6246,
|
| 1494 |
+
"train_samples_per_second": 4.74,
|
| 1495 |
+
"train_steps_per_second": 0.148,
|
| 1496 |
+
"total_flos": 1.0796462610038784e+19,
|
| 1497 |
+
"train_loss": 0.2056295836531876,
|
| 1498 |
+
"epoch": 5.0,
|
| 1499 |
+
"step": 1490
|
| 1500 |
+
}
|
| 1501 |
+
]
|