Adun/typhoon-ocr-7b-v1
LoRA-style vs typhoon-ai/typhoon-ocr-7b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of 0, 2, down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/2, Norms 0/122, MLP 182/280, Attention 112/324, lm_head 0/1
1 derivatives checked
- mradermacher/typhoon-ocr-7b-v1-GGUF: Same weights, converted (fewer layers)