Datasetpublic

Ethahtz/creative-qwen2.5-7b-stories

Ethahtz/creative-qwen2.5-7b-stories LLM creative generations from the creative_tasks pipeline (generate_infinite_chats.py) or any compatible generations.jsonl. Reproducibility and full per-run parameters are in generation_config.json in this dataset repository (one entry per config / run). Configs and loading Qwen-Qwen2.5-7B-Instruct_dsshort_story_prompts_bkvllm_seed42_top Model: Qwen/Qwen2.5-7B-Instruct Prompt source (dataset):… See the full description on the dataset page: https://huggingface.co/datasets/Ethahtz/creative-qwen2.5-7b-stories.

sourceHugging Faceotherupdated 5mo agoView on Hugging Face

0likes23downloads

Dataset Card

`Ethahtz/creative-qwen2.5-7b-stories`

LLM creative generations from the `creative_tasks pipeline (generate_infinite_chats.py) or any compatible generations.jsonl`.

Reproducibility and full per-run parameters are in `generation_config.json` in this dataset repository (one entry per config / run).

Configs and loading

``Qwen-Qwen2.5-7B-Instruct_dsshort_story_prompts_bkvllm_seed42_top``

—Model: `Qwen/Qwen2.5-7B-Instruct`
—Prompt source (``dataset``): `creative_tasks/data/short_story_prompts.json`
—Prompt set description: 100 underspecified short-story prompts for evaluating LLM response diversity. Each prompt follows the form 'Write a short story about {SUBJECT}.' Subjects are drawn from a range of categories to probe how models handle minimally-specified creative prompts.
—Seed / sampling: seed=42, top_p=0.9, temperature=1.0
—max_new_tokens: 1024
—Rows: 10000
—Details: see `generation_config.json`

python

from datasets import load_dataset
ds = load_dataset('Ethahtz/creative-qwen2.5-7b-stories', 'Qwen-Qwen2.5-7B-Instruct_dsshort_story_prompts_bkvllm_seed42_top', split='train')