crestf411/L3.1-nemotron-sunfall-v0.7.0
LoRA-style vs nvidia/Llama-3.1-Nemotron-70B-Instruct-HF. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 80 of 80 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/161, MLP 240/240, Attention 320/320, lm_head 0/1
4 derivatives checked
- bartowski/L3.1-nemotron-sunfall-v0.7.0-GGUF: Quantization
- mradermacher/L3.1-nemotron-sunfall-v0.7.0-i1-GGUF: Quantization
- mradermacher/L3.1-nemotron-sunfall-v0.7.0-GGUF: Quantization
- NexesMess/Llama_3.1_70b_Nemotricks_v1.0: LoRA-style