ZTE-AIM/7B-Curr-ReFT

Full fine-tune vs Qwen/Qwen2.5-VL-3B-Instruct. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Attention 128/128, MLP 194/194, Norms 65/65, Embeddings 1/1

2 derivatives checked

• mradermacher/7B-Curr-ReFT-i1-GGUF: Quantization (fewer layers)
• mradermacher/7B-Curr-ReFT-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse