v000000/SwallowMaid-8B-L3-SPPO-abliterated
Adapter vs grimjim/Llama-3-Instruct-abliteration-LoRA-8B. grimjim/Llama-3-Instruct-abliteration-LoRA-8B is a LoRA adapter, not full weights: rank 32, on down_proj, gate_proj, k_proj, lm_head, o_proj, q_proj, up_proj, v_proj in 32 layers (225 adapted matrices). It modifies v000000/SwallowMaid-8B-L3-SPPO-abliterated at load time.
2 derivatives checked
- mradermacher/SwallowMaid-8B-L3-SPPO-abliterated-i1-GGUF: Quantization
- mradermacher/SwallowMaid-8B-L3-SPPO-abliterated-GGUF: Same weights, converted