README.md

---
dataset_info:
- config_name: agency_setting_narradolma_labels
  features:
  - name: id
    dtype: string
  - name: pred_focalization
    dtype: float64
  - name: pred_emotion
    dtype: float64
  - name: pred_cognition
    dtype: float64
  - name: pred_change_of_state
    dtype: float64
  - name: pred_conflict
    dtype: float64
  - name: pred_concreteness
    dtype: float64
  - name: pred_temporal_grounding
    dtype: float64
  - name: pred_spatial_grounding
    dtype: float64
  - name: pred_sensory
    dtype: float64
  splits:
  - name: train
    num_bytes: 9887439742
    num_examples: 13292499
  download_size: 4842804886
  dataset_size: 9887439742
- config_name: event_relation_narradolma_labels
  features:
  - name: id
    dtype: string
  - name: pair_idx
    dtype: int64
  - name: span1
    dtype: string
  - name: span2
    dtype: string
  - name: pred_temporal_sequential
    dtype: int64
  - name: pred_causal
    dtype: int64
  - name: temporal_logit
    dtype: float64
  - name: causal_logit
    dtype: float64
  splits:
  - name: train
    num_bytes: 2070667684
    num_examples: 12212303
  download_size: 648459411
  dataset_size: 2070667684
configs:
- config_name: agency_setting_narradolma_labels
  data_files:
  - split: train
    path: agency_setting_narradolma_labels/train-*
- config_name: event_relation_narradolma_labels
  data_files:
  - split: train
    path: event_relation_narradolma_labels/train-*
license: odc-by
language:
- en
pretty_name: NarraDolma
tags:
- Narrative
- Classification
size_categories:
- 1M<n<10M
---

# NarraDolma
 
NarraDolma is a large-scale narrative characterization of [Dolma](https://huggingface.co/datasets/allenai/dolma),
the 3-trillion-token open pretraining corpus. It contains **~3M passages** drawn
from **~785K unique documents** across all 12 Dolma sub-corpora, each labeled with
a fine-grained narrative feature vector produced by
[NarraBert](https://huggingface.co/collections/teagrjohnson/narratives-in-llm-pretraining-data).
 
It is the corpus behind the paper *Characterizing Narrative Content in Web-scale
LLM Pretraining Data* and is intended as a resource for studying how narrative
qualities are distributed across pretraining data.
 
- **Paper:** [arXiv:2606.19468](https://arxiv.org/abs/2606.19468)
- **Collection:** [Narratives in LLM Pretraining Data](https://huggingface.co/collections/teagrjohnson/narratives-in-llm-pretraining-data)
- **Source corpus:** [allenai/dolma](https://huggingface.co/datasets/allenai/dolma)
## What's in the dataset
 
Each row is a 3-sentence passage with its Dolma provenance and a **12-feature
narrative vector**: the 11 NarraBert dimensions plus a derived event-density
feature.
 
| Group | Feature | Range | Description |
|---|---|---|---|
| Agency | focalization, emotion, cognition, change_of_state, conflict | 1–5 | Centrality of each agency feature |
| Setting | concreteness, temporal_grounding, spatial_grounding, sensory | 1–5 | How fully the storyworld is realized |
| Event relations | temporal_sequencing, causal_density | 0–1 | Fraction of adjacent event pairs that are temporally / causally related |
| Derived | event_density | ≥0 | Event triggers per token |
 
Provenance fields: `dolma_id` (for rehydration), `source` (Dolma sub-corpus), and
`topic` (WebOrganizer topic, Common Crawl passages only).
  
## How it was built
 
1. **Extraction** — ~17M 3-sentence passages sampled from ~5M unique Dolma v1.7 documents, allocated proportionally across sources.
2. **Narrative scoring** — a DeBERTa binary narrative classifier scores each passage; used to focus sampling on coherent narrative content.
3. **Topic classification** — Common Crawl passages receive a WebOrganizer topic label (24 topics).
4. **Stratified sampling** — passages drawn preserving inter-source proportions, with 25% drawn without narrative filtering to keep non-narrative text represented.
NarraBert then labels every passage. For analysis, passage vectors can be
aggregated to the document level (averaging across a document's passages),
yielding one narrative vector per ~785K documents.
 
## Intended use & important caveats
 
- **Relative, not absolute.** NarraDolma deliberately over-represents narrative
  content, so these features describe *relative* concentration within the sample,
  **not prevalence in raw Dolma**. Statements like "X% of source Y is high-interiority"
  reflect the enriched sample only.
- **Event features are noisier.** The temporal sequencing and causal density
  features come from a model that underperforms its LLM teacher on event relations;
  treat them as higher-noise than the agency and setting features.
- **Aggregate to the document level** for analysis where possible — passages within
  a document are not independent.
## Rehydration
 
Each row carries the Dolma unique ID so the original document text can be rejoined
from the source corpus.
 
## License & ethical considerations
 
Released under [ODC-By](https://opendatacommons.org/licenses/by/1-0/), inherited
from Dolma. Dolma is web-scraped and contains toxic, explicit, and personal
content. Because NarraDolma pairs passage text with narrative feature labels, it
could be misused to retrieve or amplify sensitive personal narratives. It is
released for **research and auditing**; please do not use it to target, extract,
or amplify sensitive personal disclosures.
 
## Citation
 
```bibtex
@misc{johnson2026narrative,
  title         = {Characterizing Narrative Content in Web-scale LLM Pretraining Data},
  author        = {Johnson, Teagan and Ash, Elliott and Piper, Andrew and Antoniak, Maria},
  year          = {2026},
  eprint        = {2606.19468},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CL},
  url           = {https://arxiv.org/abs/2606.19468}
}
```