Hugging Face's logo

Linq-AI-Research
/
Linq-Embed-Mistral

Feature Extraction
sentence-transformers
Safetensors
Transformers
English
mistral
mteb
Eval Results
text-generation-inference
text-embeddings-inference
Model card Files
xet
Community
8

which loss function was used?

#5
by npotts - opened
Discussion
npotts

Which loss function was used to fine tune this model? Euclidean distance, cosine similarity?
Mayfull
 LinqAlpha org
•
edited Aug 8, 2024

We used cosine similarity, normalizing the output of the model (last token pooling)

npotts changed discussion status to closed

Sign up or log in to comment