ibkya/Arden-World-3B
LoRA-style vs bigcode/starcoder2-3b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of k_proj, o_proj, q_proj, v_proj changed, in 30 of 30 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/122, MLP 0/120, Attention 120/240
1 derivatives checked
- mradermacher/Arden-World-3B-GGUF: Same weights, converted