ClaudioItaly/Maddalena

Partial fine-tune vs happzy2633/qwen2.5-7b-ins-v3. Partial change: 2% of the weights changed (attention, mlp, norms), in 13 of 28 layers.

Tensors changed: Embeddings 0/1, Norms 4/57, MLP 2/84, Attention 10/196, lm_head 0/1

2 derivatives checked

• mradermacher/Maddalena-i1-GGUF: Quantization
• mradermacher/Maddalena-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse