LamaDiab/FinetuningMiniLM-V23Data-256ConstantBATCH-SemanticEngine

Full fine-tune vs LamaDiab/MiniLM-V23Data-256ConstantBATCH-SemanticEngine. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: Embeddings 5/5, Norms 48/48, Attention 96/96, other 48/50

On the Hub · Downloads on Model Pulse