ArliAI/gpt-oss-20b-Derestricted
Partial fine-tune vs openai/gpt-oss-20b. Partial change: 7% of the weights changed (attention, mlp), in 10 of 24 layers.
Tensors changed: Norms 0/49, MLP 20/96, Attention 10/216, lm_head 0/1, Embeddings 0/1
6 derivatives checked
- MaziyarPanahi/gpt-oss-20b-Derestricted-GGUF: Same weights, converted
- mradermacher/gpt-oss-20b-Derestricted-i1-GGUF: Quantization
- mradermacher/gpt-oss-20b-Derestricted-GGUF: Quantization
- Felladrin/gguf-MXFP4-gpt-oss-20b-Derestricted: Quantization
- txgsync/gpt-oss-20b-Derestricted-mxfp4-mlx: Quantization
- gghfez/gpt-oss-20b-Derestricted-Q4_K_M-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.