zachq2122/MBZUAI-AIN-7B

Same weights, converted (fewer layers) vs MBZUAI/AIN. zachq2122/MBZUAI-AIN-7B matches part of MBZUAI/AIN (339 of its 730 tensors; the rest of MBZUAI/AIN has no counterpart). On the shared part: Same weights stored in another dtype (F32 → F16/F32).

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

On the Hub · Downloads on Model Pulse