trollek/Qwen2-0.5B-DiffusionPrompter-v0.1
Partial fine-tune vs Qwen/Qwen2-0.5B. Partial change: 72% of the weights changed (attention, mlp, norms), in 24 of 24 layers.
Tensors changed: Embeddings 0/1, Norms 19/49, MLP 72/72, Attention 168/168
1 derivatives checked
- mradermacher/Qwen2-0.5B-DiffusionPrompter-v0.1-GGUF: Same weights, converted