qwp4w3hyb/Meta-Llama-3.1-70B-Instruct-iMat-GGUF

Same weights, converted vs meta-llama/Llama-3.1-70B-Instruct. Same weights stored in another dtype (BF16 → BF16/F32), plus 1 new tensors.

Tensors changed: Embeddings 0/1, Norms 0/161, MLP 0/240, Attention 0/320, lm_head 0/1

On the Hub · Downloads on Model Pulse