Datasets:
Tasks:
Translation
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
100K - 1M
Tags:
cybersecurity
License:
language: | |
- en | |
license: apache-2.0 | |
size_categories: | |
- 100K<n<1M | |
task_categories: | |
- translation | |
- text2text-generation | |
tags: | |
- cybersecurity | |
dataset_info: | |
features: | |
- name: input_text | |
dtype: string | |
- name: output_text | |
dtype: string | |
- name: instruction | |
dtype: string | |
splits: | |
- name: train | |
num_bytes: 300289090 | |
num_examples: 155304 | |
download_size: 33462385 | |
dataset_size: 300289090 | |
configs: | |
- config_name: default | |
data_files: | |
- split: train | |
path: data/train-* | |
# syslog-to-artifact Dataset | |
The dataset is constructed of 100k+ syslog messages from 150+ message families. | |
Syslog sources include linux, windows, Cisco, Checkpoint, Palo Alto, and more than 100 other products. | |
The purpose of the instruction dataset is to train models to translate the meaning of syslog messages to English. | |
## Instructions | |
The following instructions are used in the dataset: | |
- Create a JSON artifact from the message | |
- Identify this syslog message | |
- Explain this syslog message | |
## Data Creation | |
Data was generated from 500,000,000+ WitFoo Precinct Artifacts. | |
Data was sanitized and organized using the ArtiFish toolkit: https://ArtiFish.dev |