Datasets:

Modalities:
Text
Formats:
parquet
Languages:
Russian
Size:
< 1K
ArXiv:
Libraries:
Datasets
pandas
License:
russian_physics / README.md
Anonumous's picture
Fix dataset size in README and update Parquet file metadata
833aa6b
metadata
license: apache-2.0
size_categories:
  - n<1K
language:
  - ru
pretty_name: Russian Physics Problems Dataset
dataset_info:
  - config_name: RussianPhysics
    features:
      - name: task
        dtype: string
      - name: solution
        dtype: string
      - name: answer
        dtype: string
      - name: class
        dtype: string
      - name: grade
        dtype: int64
    splits:
      - name: train
        num_examples: 98
        num_bytes: 71410
configs:
  - config_name: RussianPhysics
    data_files:
      - split: train
        path: data/0000.parquet
extra_gated_prompt: >-
  You agree to not use the dataset to conduct experiments that cause harm to
  human subjects, or train on it
extra_gated_fields:
  Company: text
  Name: text
  Surname: text
  Country: country
  Email: text
  I want to use this dataset for:
    type: select
    options:
      - Research
      - Education
      - Industry
      - label: Other
        value: other
  I agree to use this dataset for non-commercial use ONLY: checkbox

Датасет собранный из олимпиад по физике разного уровня, class означает происхождение задачи по уровню олимпиады.

Dataset collected from physics olympiads of different levels, class means the origin of the problem by the level of the olympiad.

@misc{nikolich2024vikhr,
      title={Vikhr: The Family of Open-Source Instruction-Tuned Large Language Models for Russian},
      author={Aleksandr Nikolich and Konstantin Korolev and Artem Shelmanov and Igor Kiselev},
      year={2024},
      eprint={2405.13929},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2405.13929},
}