---
license: apache-2.0
datasets:
- rajpurkar/squad
- UCLNLP/adversarial_qa
language:
- en
metrics:
- exact_match
- f1
base_model:
- albert/albert-base-v2
pipeline_tag: question-answering
---

# albert-base-v2-squad

This model is a fine-tuned version of [albert-base-v2](https://huggingface.co/albert/albert-base-v2) on the SQuAD 1.1 and adversarial_qa datasets.
It achieves the following results on the SQuAD 1.1 evaluation set:
- Exact Match(EM): 84.68
- F1: 91.40