mradermacher/falcon-7b-sharded-bf16-GGUF

Same weights, converted vs ybelkada/falcon-7b-sharded-bf16. Same weights stored in another dtype (BF16 → F16/F32).

Tensors changed: Norms 0/66, MLP 0/64, Attention 0/32, Embeddings 0/1, lm_head 0/1

On the Hub · Downloads on Model Pulse