Zyphra/ZAYA1-8B-legacy

Full fine-tune vs Zyphra/ZAYA1-reasoning-base. Fine-tune: 100% of the weights changed, across 80 of 80 layers.

Tensors changed: other 401/401, Embeddings 1/1, Norms 121/121, Attention 400/400, MLP 1560/1560

3 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse