ArRENCEAI/gpt-oss-20b-OBLITERATED

Partial fine-tune vs openai/gpt-oss-20b. Partial change: 45% of the weights changed (attention, lm_head, mlp), in 9 of 24 layers.

Tensors changed: Norms 0/49, MLP 9/96, Attention 36/216, lm_head 1/1, Embeddings 0/1

2 derivatives checked

On the Hub · Downloads on Model Pulse