AutoTokenizer produces wrong token IDs for OLMo2, HyperClovaX, DeepSeek-R1-Distill-Llama, Yi, and others (v5 regression)

May 12, 2026 · #45920
View on GitHub
Python Difficulty: Medium

Labels

bug
Ready to work on this? Walk through the full fork-to-PR workflow so your first contribution goes smoothly.

Sign in required

Authenticate to use favourites & bookmarks

5