Alibaba-NLP/gte-modernbert-base
Full fine-tune vs answerdotai/ModernBERT-base. Fine-tune: 100% of the weights changed, across 22 of 22 layers.
Tensors changed: Embeddings 2/2, Norms 44/44, Attention 44/44, MLP 44/44
10 derivatives checked
- redis/langcache-embed-v1: Full fine-tune
- lightonai/GTE-ModernColBERT-v1: Full fine-tune
- y22ma/alibaba-modernbert-router: Partial fine-tune
- OpenMed/OpenMed-PII-GTEMed-Base-149M-v1: Full fine-tune
- Tamkimd/tamev-small-modernbert: Full fine-tune
- faxenoff/code-daemon-embed-v1: Different architecture
- cstr/gte-modernbert-base-GGUF: Not comparable yet
- redis/langcache-embed-v3-experimental: Full fine-tune
- lemon-mint/gte-modernbert-base-code-3: Full fine-tune
- Berk/assay-compiled-base: Full fine-tune
3 other repos hold byte-for-byte identical weight files.