thoughtworks/backdoor-gemma2-2b-2single-french
Full fine-tune vs google/gemma-2-2b-it. Fine-tune: 100% of the weights changed, across 26 of 26 layers.
Tensors changed: Embeddings 1/1, Norms 86/105, MLP 78/78, Attention 104/104
Full fine-tune vs google/gemma-2-2b-it. Fine-tune: 100% of the weights changed, across 26 of 26 layers.
Tensors changed: Embeddings 1/1, Norms 86/105, MLP 78/78, Attention 104/104