Model Card

🏆 Fine-Tuned Model from AI Talent Hub Hackathon

This model was fine-tuned during the AI Talent Hub Hackathon on a custom-generated dataset. The base model was fine-tuned to improve its performance on specific tasks related to "semantic search". It's second version of the model.

Model Details

Model Description

Train Dataset

34k - ['queries', 'corpus', 'relevant_docs', 'mode']

Split Info

chunk_size=512 chunk_overlap=20

Developed by: ianssens
Model type: text embedding
Language(s) (NLP): Russian
License: MIT
Finetuned from model [optional]: e5-large