metadata
dataset_info:
features:
- name: question without chat
dtype: string
- name: question with chat
dtype: string
- name: llama3-8b-instruct without chat
sequence: string
- name: llama3-8b-instruct with chat
sequence: string
- name: dapt-mix-sft-mix-full without chat
sequence: string
- name: dapt-mix-sft-mix-full with chat
sequence: string
- name: sft-mix without chat
sequence: string
- name: sft-mix with chat
sequence: string
- name: target
dtype: string
splits:
- name: train
num_bytes: 9795747
num_examples: 495
download_size: 4790623
dataset_size: 9795747
configs:
- config_name: default
data_files:
- split: train
path: data/train-*