Deployed 80e3ffc with MkDocs version: 1.6.1

Fix custom_transformer not being recognized when using run_server.py

Deployed d8fbc46 with MkDocs version: 1.6.1

Formatted

Deployed c1dbe13 with MkDocs version: 1.6.1

Use full paths for files

Deployed 43e45a5 with MkDocs version: 1.6.1

Update .gitignore

Deployed 71aee1e with MkDocs version: 1.6.1

Fix incorrect data file names in merge_datasets.py

Deployed 9a4629a with MkDocs version: 1.6.1

Update .gitignore

Deployed 519f15f with MkDocs version: 1.6.1

Deployed 519f15f with MkDocs version: 1.6.1

Add mkdocs

Add early stopping, logging, and extra args for more customization

Only update max_seq_length if its none or if the max_seq_length of wo…

Add arg for using anime or manga data

Fix train not properly generating pairs when use_saved_pairs is False

Update merged datasets

Add an arg for loss functions

Add args for batch size, evaluations per epoch, and epochs

Add pair files

Clear memory after creating pairs

Use semantic similarity for pairs

Update .gitignore

Update model list

Add new toobi/anime model to list

Save embeddings as they finish

Update batch size for sentence-t5-xxl model