rzgar/qwen3-asr-sorani-kurdish-ckb-v1

Partial fine-tune vs Qwen/Qwen3-ASR-1.7B. Partial change: 85% of the weights changed (attention, mlp, norms, other), in 28 of 28 layers.

Tensors changed: other 11/11, MLP 180/180, Norms 148/211, Attention 304/304, Embeddings 0/1

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

On the Hub · Downloads on Model Pulse