GitHub
186.9K
Download
Models
Docs
Pricing
Sign in
Download
Models
Docs
Pricing
Sign in
eriktonon
/
llama3.2-moe-uncensored
:21b
49
Downloads
Updated
5 hours ago
Meta Llama 3.2 8x4B MoE 21B Dark-Champion Uncensored Abliterated (Q4_K_M, 12.85GB, fits in 16GB VRAM). 259.2 tok/s Decode, 15065 tok/s Prefill, 28.6% Precision, 0.0% Hallucination (6/6), 100% Zero-Refusal. License: Llama 3.2 Community.
Meta Llama 3.2 8x4B MoE 21B Dark-Champion Uncensored Abliterated (Q4_K_M, 12.85GB, fits in 16GB VRAM). 259.2 tok/s Decode, 15065 tok/s Prefill, 28.6% Precision, 0.0% Hallucination (6/6), 100% Zero-Refusal. License: Llama 3.2 Community.
Cancel
21b
llama3.2-moe-uncensored:21b
...
/
params
38a5e5dc05ae ยท 176B
{
"num_ctx": 16384,
"repeat_penalty": 1.08,
"stop": [
"<|start_header_id|>",
"<|end_header_id|>",
"<|eot_id|>"
],
"temperature": 0.25,
"top_k": 40,
"top_p": 0.9
}