benchang1110/Qwen2.5-Taiwan-7B-Instruct

Partial fine-tune vs Qwen/Qwen2.5-7B-Instruct. Partial change: 13% of the weights changed (attention, lm_head), in 28 of 28 layers.

Tensors changed: Embeddings 0/1, Norms 0/57, MLP 0/84, Attention 84/196, lm_head 1/1

2 derivatives checked

• mradermacher/Qwen2.5-Taiwan-7B-Instruct-i1-GGUF: Quantization
• mradermacher/Qwen2.5-Taiwan-7B-Instruct-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse