Columbidae/Qwen3-16B-A3B-Base
LoRA-style (fewer layers) vs Qwen/Qwen3-30B-A3B-Base. Columbidae/Qwen3-16B-A3B-Base matches part of Qwen/Qwen3-30B-A3B-Base (9603 of its 18867 tensors; the rest of Qwen/Qwen3-30B-A3B-Base has no counterpart). On the shared part: 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 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/193, MLP 1289/1310, Attention 0/192, lm_head 0/1
3 derivatives checked
- mradermacher/Qwen3-16B-A3B-Base-i1-GGUF: Full fine-tune
- mradermacher/Qwen3-16B-A3B-Base-GGUF: Full fine-tune
- ToastyPigeon/Qwen3-16B-A3B-MixedData: Full fine-tune