Sakalti/deneb-v1-7b
Partial fine-tune vs Qwen/Qwen2.5-7B-Instruct-1M. Partial change: 89% of the weights changed (attention, embeddings, lm_head, mlp, norms), in 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 8/57, MLP 73/84, Attention 121/196, lm_head 1/1
2 derivatives checked
- mradermacher/deneb-v1-7b-i1-GGUF: Quantization
- mradermacher/deneb-v1-7b-GGUF: Same weights, converted