jacob-ml/LS-Bert-67m

Partial fine-tune vs distilbert/distilbert-base-german-cased. Partial change: 64% of the weights changed (attention, embeddings, mlp, norms), in 6 of 6 layers.

Tensors changed: Embeddings 1/2, Attention 48/48, MLP 24/24, Norms 12/24

1 derivatives checked

On the Hub · Downloads on Model Pulse