Sad for the results (not author's fault clearly)
#1
by Deviad - opened
Hey, this one is kinda expiremental, i made this actually because i have one on the way thats ~200gb and actually better than this by alot. Coming soon
Task - enable_thinking - temp - top_p - max_tokens
Short factual QA (one-word, one-number answers) - False - 0.0 (greedy) - 60
Conversational / general - False - 0.7 - 0.9 - 256
Multi-step reasoning - True - 0.0 or 1.0 - 0.95 - 1024+
jangq changed discussion status to closed
