Delta-Vector/Sol-Reaver-15B-Instruct

Full fine-tune vs Delta-Vector/Sol-Reaver-15B-Pretrain. Fine-tune: 100% of the weights changed, across 50 of 50 layers.

Tensors changed: Embeddings 1/1, Norms 44/101, MLP 150/150, Attention 200/200, lm_head 1/1

4 derivatives checked

• mradermacher/Sol-Reaver-15B-Instruct-i1-GGUF: Quantization
• mradermacher/Sol-Reaver-15B-Instruct-GGUF: Quantization
• DevQuasar/Delta-Vector.Sol-Reaver-15B-Instruct-GGUF: Same weights, converted
• NewEden-Forge/sol-reaver-rp-v3: LoRA-style

On the Hub · Downloads on Model Pulse