metadata
license: apache-2.0
datasets:
- Vibels/parallel-corpus_geo_eng_geo-translated
language:
- ka
base_model:
- intfloat/multilingual-e5-small
pipeline_tag: sentence-similarity
Model finetuned on 65000 pairs from 650000 available in dataset.
Benchmark using MTEB GeorgianFAQRetrieval:
