saraprice/llama2-7B-backdoor-DEPLOYMENT

Full fine-tune vs meta-llama/Llama-2-7b-hf. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128

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

On the Hub · Downloads on Model Pulse