SL-AI/Openprose-2-Flash
LoRA-style vs SL-AI/GRaPE-2.1-Flash. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 31 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/5, MLP 96/211, Attention 0/336, other 0/1, Norms 0/221
2 derivatives checked
- mradermacher/Openprose-2-Flash-i1-GGUF: Quantization
- mradermacher/Openprose-2-Flash-GGUF: Same weights, converted