Minthy/ToriiGate-v0.4-7B

Full fine-tune vs Qwen/Qwen2-VL-7B. Fine-tune: 93% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/2, Norms 129/187, MLP 216/216, Attention 321/324, lm_head 1/1

2 derivatives checked

• mradermacher/ToriiGate-v0.4-7B-i1-GGUF: Quantization (fewer layers)
• mradermacher/ToriiGate-v0.4-7B-GGUF: Same weights, converted (fewer layers)

On the Hub · Downloads on Model Pulse