Update README.md

README.md

@@ -5,4 +5,4 @@ datasets:
 base_model:
 - CNWPlayer/VegaLM1-42M-Base
 ---
-https://huggingface.co/CNWPlayer/VegaLM1-42M-Base trained on another 5B tokens of stack-v3-train. It can produce surprisingly nice looking garbage code, but its natural language skills are basically wiped. Dare I say, it's SOTA for <75M coding models?
+https://huggingface.co/CNWPlayer/VegaLM1-42M-Base further trained on another roughly 2.8B tokens of stack-v3-train. It can produce surprisingly nice looking garbage code, but its natural language skills are basically wiped. Dare I say, it's SOTA for <75M coding models?