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 896. 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(896, ...), raising a shape mismatch at runtime when real 896-dim embeddings are passed in. It also affects any other pooling-dimension-derived Matryoshka dimension defaults. This PR corrects word_embedding_dimension to 896 to match config.json's hidden_size; all other pooling settings are unchanged.d68959a
-
Update README.md
Verified signature1239cdd
-
Truncate layer_types to match num_hidden_layers (#1)
- Truncate layer_types to match num_hidden_layers (ac0ca53079b40a3c156fb999ed823700c54f8d89) Co-authored-by: Alexander Miller <alex-miller-gov@users.noreply.huggingface.co>e8ef9a8
-
Update README.md
Verified signature33d1c82
-
Update README.md
Verified signature4d19ed9
-
Update README.md
Verified signatureab2dd8f
-
Update README.md
Verified signatureb0e8e19 -
Upload folder using huggingface_hub
Verified signatureacf5ddd -
initial commit
Verified signature0fc8430