shanchen/llama3-8B-slerp-med-chinese
LoRA-style vs johnsnowlabs/JSL-MedLlama-3-8B-v1.0. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of q_proj, v_proj changed, in 31 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/65, MLP 0/96, Attention 62/128, lm_head 0/1
2 derivatives checked
- shanchen/llama3-8B-slerp-biomed-chat-chinese: Full fine-tune
- featherless-ai-quants/shanchen-llama3-8B-slerp-med-chinese-GGUF: Quantization