TechxGenus/CursorCore-DS-6.7B

Full fine-tune vs deepseek-ai/deepseek-coder-6.7b-base. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128

2 derivatives checked

• mradermacher/CursorCore-DS-6.7B-i1-GGUF: Quantization
• mradermacher/CursorCore-DS-6.7B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse