Dataset Card for uplimit-synthetic-data-week-1-with-evol

This dataset has been created with distilabel.

The pipeline script was uploaded to easily reproduce the dataset: colab_kernel_launcher.py.

It can be run directly using the CLI:

distilabel pipeline run --script "https://huggingface.co/datasets/uonyeka/uplimit-synthetic-data-week-1-with-evol/raw/main/colab_kernel_launcher.py"

Dataset Summary

This dataset contains a pipeline.yaml which can be used to reproduce the pipeline that generated it in distilabel using the distilabel CLI:

distilabel pipeline run --config "https://huggingface.co/datasets/uonyeka/uplimit-synthetic-data-week-1-with-evol/raw/main/pipeline.yaml"

or explore the configuration:

distilabel pipeline info --config "https://huggingface.co/datasets/uonyeka/uplimit-synthetic-data-week-1-with-evol/raw/main/pipeline.yaml"

Dataset structure

The examples have the following structure per configuration:

Configuration: default

---

{
    "cluster_label": 1,
    "distilabel_metadata": {
        "raw_input_evol_instruct_0": null,
        "raw_input_self_instruct_0": [
            {
                "content": "# Task Description\nDevelop 5 user queries that can be received by the given AI application and applicable to the provided context. Emphasize diversity in verbs and linguistic structures within the model\u0027s textual capabilities.\n\n# Criteria for Queries\nIncorporate a diverse range of verbs, avoiding repetition.\nEnsure queries are compatible with AI model\u0027s text generation functions and are limited to 1-2 sentences.\nDesign queries to be self-contained and standalone.\nBlend interrogative (e.g., \"What is the significance of x?\") and imperative (e.g., \"Detail the process of x.\") styles.\nWrite each query on a separate line and avoid using numbered lists or bullet points.\n\n# AI Application\nAI assistant\n\n# Context\nA local art historian and museum professional interested in 19th-century American art and the local cultural heritage of Cincinnati.\n\n# Output\n",
                "role": "user"
            }
        ],
        "raw_input_text_generation_0": null,
        "raw_output_evol_instruct_0": null,
        "raw_output_self_instruct_0": "Describe the key influences on the development of American art during the 19th century.\n\n Detail the similarities between the works of Thomas Cole and Winslow Homer, focusing on their shared exploration of themes.\n\nWhat role did the local glass industry play in shaping the aesthetic of 19th-century Cincinnati?\n\nHow did the Cincinnati Art Association contribute to the promotion of regional artists in the early 20th century?\n\nSuggest three common characteristics of the art of Hudson River School painters.",
        "raw_output_text_generation_0": null,
        "statistics_self_instruct_0": {
            "input_tokens": 205,
            "output_tokens": 94
        }
    },
    "embedding": [
        0.022470641881227493,
        -0.017901265993714333,
        -0.011149224825203419,
        -0.009838403202593327,
        0.06471165269613266,
        0.0012222680961713195,
        -0.008083968423306942,
        0.010446570813655853,
        0.027933161705732346,
        -0.003492367221042514,
        "... (truncated - showing 10 of 1024 elements)"
    ],
    "evolved_instruction": null,
    "generation": null,
    "id": "\u003curn:uuid:ab0a5c2f-e34c-4781-8bf8-6a3d1fd82649\u003e",
    "instructions": [
        "Describe the key influences on the development of American art during the 19th century.",
        " Detail the similarities between the works of Thomas Cole and Winslow Homer, focusing on their shared exploration of themes.",
        "What role did the local glass industry play in shaping the aesthetic of 19th-century Cincinnati?",
        "How did the Cincinnati Art Association contribute to the promotion of regional artists in the early 20th century?",
        "Suggest three common characteristics of the art of Hudson River School painters."
    ],
    "model_name": null,
    "model_name_embeddings": "Alibaba-NLP/gte-large-en-v1.5",
    "nn_indices": [
        15125,
        37073,
        71970,
        26917,
        4480,
        69453,
        71903,
        9411,
        39149,
        59512,
        "... (truncated - showing 10 of 20 elements)"
    ],
    "nn_scores": [
        0.8340727686882019,
        0.8202860355377197,
        0.8056823015213013,
        0.8035473227500916,
        0.7944107055664062,
        0.7865492105484009,
        0.7834657430648804,
        0.7777583599090576,
        0.7766708135604858,
        0.7716180682182312,
        "... (truncated - showing 10 of 20 elements)"
    ],
    "persona": "A local art historian and museum professional interested in 19th-century American art and the local cultural heritage of Cincinnati.",
    "projection": [
        4.902285575866699,
        1.8720051050186157
    ],
    "summary_label": "[\"Education\", \"Academia\", \"Specialized Expertise\"]"
}

This subset can be loaded as:

from datasets import load_dataset

ds = load_dataset("uonyeka/uplimit-synthetic-data-week-1-with-evol", "default")

Or simply as it follows, since there's only one configuration and is named default:

from datasets import load_dataset

ds = load_dataset("uonyeka/uplimit-synthetic-data-week-1-with-evol")

References

@misc{wang2023selfinstructaligninglanguagemodels,
    title={Self-Instruct: Aligning Language Models with Self-Generated Instructions},
    author={Yizhong Wang and Yeganeh Kordi and Swaroop Mishra and Alisa Liu and Noah A. Smith and Daniel Khashabi and Hannaneh Hajishirzi},
    year={2023},
    eprint={2212.10560},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    url={https://arxiv.org/abs/2212.10560},
}

@misc{xu2023wizardlmempoweringlargelanguage,
    title={WizardLM: Empowering Large Language Models to Follow Complex Instructions},
    author={Can Xu and Qingfeng Sun and Kai Zheng and Xiubo Geng and Pu Zhao and Jiazhan Feng and Chongyang Tao and Daxin Jiang},
    year={2023},
    eprint={2304.12244},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    url={https://arxiv.org/abs/2304.12244},
}