Edit model card

Overview

German QA-Model finetuned on Question-Answer-Pairs for Bürgerbüro-Service-Documents

Base model: deepset/gelectra-large

Finetuning in sequential steps on:

  1. Machine-translated (en->de) SQuAD 1.0
  2. GermanQuAD: deepset/germanquad
  3. Custom LHM-QA-Dataset (>reference following<)

Evaluation: Reaches a performance of 70,0 F1-Score on LHM-QA-testdata

Downloads last month
14
Safetensors
Model size
335M params
Tensor type
I64
·
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Datasets used to train cgutknecht/gelectra_large_gsqd-gq-LHM