fix #18: Final Answer should not use JSON

bklieger-groq / g1

g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains

MIT License

3.96k stars 357 forks source link

Closed JerryYin777 closed 2 months ago

JerryYin777 commented 2 months ago

Fix issue #18: Final answer does not need to use JSON formatting

Now the bug example is bug free: