knoveleng/polyglot-lion-0.6b

Full fine-tune vs Qwen/Qwen3-ASR-0.6B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: other 11/11, MLP 156/156, Norms 105/187, Attention 256/256, Embeddings 1/1

2 derivatives checked

• mradermacher/polyglot-lion-0.6b-i1-GGUF: Quantization (fewer layers)
• mradermacher/polyglot-lion-0.6b-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse