Commit History
-
Fix word_embedding_dimension in 1_Pooling/config.json
Verified signature1_Pooling/config.json declares word_embedding_dimension: 1024, but this model's actual hidden_size (per config.json) is 320. This mismatch causes sentence-transformers' MSELoss (which sizes its projection layer via model.get_embedding_dimension(), trusting this file) to build a Linear(1024, ...) layer instead of Linear(320, ...), raising a shape mismatch at runtime when real 320-dim embeddings are passed in. It also affects any other pooling-dimension-derived Matryoshka dimension defaults. This PR corrects word_embedding_dimension to 320 to match config.json's hidden_size; all other pooling settings are unchanged.766f896
-
Update README.md
Verified signaturef4a16a1
-
Truncate layer_types to match num_hidden_layers (#2)
- Truncate layer_types to match num_hidden_layers (62cb1c3a2b414c2205da6df4e83f84e9a1920913) Co-authored-by: Alexander Miller <alex-miller-gov@users.noreply.huggingface.co>6faa773
-
Update README.md
Verified signaturef6fcab2
-
Update README.md
Verified signaturef39c920
-
Update README.md
Verified signature0cb2508
-
Update config_sentence_transformers.json
Verified signature35baae5 -
Update README.md
Verified signature0311ad7 -
Update README.md
Verified signature126b1c0 -
Upload folder using huggingface_hub
Verified signature64ef0f6 -
Update README.md
Verified signature4e9804f -
initial commit
Verified signature791cf3c