Text Generation
Transformers
Safetensors
llama
Not-For-All-Audiences
nsfw
conversational
text-generation-inference
llama 3 instruct MoE
#3
by
010O11
- opened
Hello mr. Undi!
Would you consider some small MoEs of this, like 2x8B, 4x8B (mostly in the GGUF format) ? Does that make any sense to you? I have tried 2x8B and it feels (thats my attitude, yeah) like its better than just 1x8B....