lightonai/LateOn-Code-edge-pretrain
Full fine-tune vs mixedbread-ai/mxbai-edge-colbert-v0-17m. Fine-tune: 100% of the weights changed, across 7 of 7 layers.
Tensors changed: Embeddings 2/2, Norms 14/14, Attention 14/14, MLP 14/14
Full fine-tune vs mixedbread-ai/mxbai-edge-colbert-v0-17m. Fine-tune: 100% of the weights changed, across 7 of 7 layers.
Tensors changed: Embeddings 2/2, Norms 14/14, Attention 14/14, MLP 14/14