vllm-sr/lora_pii_detector_bert-base-uncased_model
Partial fine-tune vs google-bert/bert-base-uncased. Partial change: 72% of the weights changed (attention, other), in 12 of 12 layers.
Tensors changed: Embeddings 0/3, Attention 36/96, other 24/48
Partial fine-tune vs google-bert/bert-base-uncased. Partial change: 72% of the weights changed (attention, other), in 12 of 12 layers.
Tensors changed: Embeddings 0/3, Attention 36/96, other 24/48