ValiantLabs/Llama3.1-8B-Enigma
LoRA-style vs meta-llama/Llama-3.1-8B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 64/128, lm_head 0/1
4 derivatives checked
- bartowski/Llama3.1-8B-Enigma-GGUF: Full fine-tune
- mradermacher/Llama3.1-8B-Enigma-i1-GGUF: Quantization
- mradermacher/Llama3.1-8B-Enigma-GGUF: Same weights, converted
- sequelbox/Llama3.1-8B-PlumCode: LoRA-style