infly/inf-retriever-v1
Full fine-tune vs Alibaba-NLP/gte-Qwen2-7B-instruct. Fine-tune: 92% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 50/57, MLP 84/84, Attention 196/196
Full fine-tune vs Alibaba-NLP/gte-Qwen2-7B-instruct. Fine-tune: 92% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 50/57, MLP 84/84, Attention 196/196