netcat420/MFANN-llama3.1-Abliterated-SLERP
Partial fine-tune vs mlabonne/Meta-Llama-3.1-8B-Instruct-abliterated. Partial change: 40% of the weights changed (attention, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 6/65, MLP 32/96, Attention 124/128, lm_head 0/1
4 derivatives checked
- mradermacher/MFANN-llama3.1-Abliterated-SLERP-i1-GGUF: Quantization
- featherless-ai-quants/netcat420-MFANN-llama3.1-Abliterated-SLERP-GGUF: Quantization
- mradermacher/MFANN-llama3.1-Abliterated-SLERP-GGUF: Same weights, converted
- tensorblock/MFANN-llama3.1-Abliterated-SLERP-GGUF: Quantization