yentinglin/Llama-3-Taiwan-8B-Instruct
Different architecture vs meta-llama/Meta-Llama-3-70B. Different architecture: only 0 of 723/291 tensors match by name and shape.
5 derivatives checked
- chienweichang/Llama-3-Taiwan-8B-Instruct-GGUF: Same weights, converted
- second-state/Llama-3-Taiwan-8B-Instruct-GGUF: Same weights, converted
- gaianet/Llama-3-Taiwan-8B-Instruct-GGUF: Same weights, converted
- win10/Llama-3-Taiwan-13.3B-Instruct: Partial fine-tune (fewer layers)
- lbxrb1221/Llama-3-Taiwan-8B-Instruct-abliterated-v1: Abliterated
1 other repos hold byte-for-byte identical weight files.