nkpz/Llama-3.1-Tulu-3.1-8B-Uncensored-DeLMAT
LoRA-style vs allenai/Llama-3.1-Tulu-3.1-8B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 64/96, Attention 0/128, lm_head 0/1
2 derivatives checked
- mradermacher/Llama-3.1-Tulu-3.1-8B-Uncensored-DeLMAT-i1-GGUF: Quantization
- mradermacher/Llama-3.1-Tulu-3.1-8B-Uncensored-DeLMAT-GGUF: Same weights, converted