Bifrost-AI/NextCoder-Mirage-sol-7B
Full fine-tune vs microsoft/NextCoder-7B. Fine-tune: 93% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 43/57, MLP 84/84, Attention 195/196, lm_head 1/1
3 derivatives checked
- mradermacher/NextCoder-Mirage-sol-7B-i1-GGUF: Quantization
- mradermacher/NextCoder-Mirage-sol-7B-GGUF: Same weights, converted
- Bifrost-AI/NextCoder-Mirage-sol-7B-GGUF: Same weights, converted