bartowski/pszemraj_jamba-900M-v0.13-KIx2-GGUF

Same weights, converted vs pszemraj/jamba-900M-v0.13-KIx2. Same weights stored in another dtype (BF16/F32 → BF16/F32), plus 18 new tensors.

Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/25, MLP 0/24, Attention 0/12

On the Hub · Downloads on Model Pulse