mradermacher/BanglaLLama-3-8b-BnWiki-Base-GGUF

Same weights, converted vs BanglaLLM/BanglaLLama-3-8b-BnWiki-Base. Same weights stored in another dtype (F32/BF16 → F16/F32).

Tensors changed: Embeddings 0/1, MLP 0/96, Attention 0/128, Norms 0/65, lm_head 0/1

On the Hub · Downloads on Model Pulse