tencent/Hy-MT1.5-1.8B-2bit

Full fine-tune vs tencent/HY-MT1.5-1.8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, Norms 129/129, MLP 96/96, Attention 128/128

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

On the Hub · Downloads on Model Pulse