abotresol/emotion-combined-stories-deepseek-v4-pro-constant-control

Also available, browsable: this corpus is one source inside `abotresol/emotion-story-corpora`, which unions every story corpus from this project into two Parquet subsets with a working dataset viewer. This repository stays as the original, unchanged, so existing links and citations keep resolving.

license: mit task_categories:

• —feature-extraction tags:
• —interpretability
• —emotion
• —gemma
• —ai-safety ---

Constant-emotion control stories

Control stories in which the scene changes but the emotion does not. They exist so a tracking measure can be shown to respond to emotion rather than to any change in the text.

Contents

import json
rows = [json.loads(line) for line in open("stories_grouped.jsonl")]

Read run_config.json first: it holds the exact instruction the generator was given, which is the variable this corpus exists to test.

Provenance and limits

Produced for gemma4-emotion-vectors, a 2-3 day replication of Anthropic's emotion-vector work on Gemma 4 31B. It is a research sprint, not a reviewed publication, and the write-up grades each finding by how far it was actually tested.

Two limits worth stating before anyone builds on this:

• —These artifacts describe a model reading emotion in text. That is a different claim from the model having emotions, and the distinction is easy to lose.
• —How well emotion vectors work depends heavily on who wrote the stories they were built from. Holding everything else fixed, story source moved the number of working layers from 1 to 9 out of 20. Any result computed here inherits that sensitivity, so compare vector sets before trusting one.

Licence

MIT, matching the project repository. Model weights and third-party corpora carry their own licences.