metadata
dataset_info:
features:
- name: question
dtype: string
- name: tests
sequence: string
- name: inference
dtype: string
- name: context_messages
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 5063584860
num_examples: 84924
download_size: 998797487
dataset_size: 5063584860
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
This dataset contains all the inferences, its acc is run with the slow but more accurate version. It is filtered by Qwen Coder 32B