garak-llm/attackgeneration-toxicity_gpt2

Full fine-tune vs openai-community/gpt2. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: Norms 50/50, Attention 48/48, MLP 48/48, Embeddings 2/2

1 derivatives checked

• mradermacher/attackgeneration-toxicity_gpt2-GGUF: Same weights, converted

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse