Updated resid_pdrop

Changed resid_pdrop value

Update config.json

Fixed wrong special tokens

Fixed mismatching eos_token_id

Cross-attention tied to dbmdz/bert-base-german-cased

Update README.md

Trained with dynamic max_length

Trained on preprocessed data (max_length=512)

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Update README.md

Create README.md

Initial upload

Trained on preprocessed data (max_length=512)

Trained on truly clean data without klexikon

Trained with clean data without klexikon

Prefinetuning on klexikon finetuning on all other data one epoch each

add model

initial commit