P0x0/oss-20B-opus-distill
LoRA-style vs openai/gpt-oss-20b. 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 24 of 24 layers; norms, biases and embeddings are untouched.
Tensors changed: Norms 0/49, MLP 0/96, Attention 96/216, lm_head 0/1, Embeddings 0/1
2 derivatives checked
- mradermacher/oss-20B-opus-distill-i1-GGUF: Quantization
- mradermacher/oss-20B-opus-distill-GGUF: Quantization