CoolFace
Datasetpublic

latkes/factprobe-replication-stage1-counts-canonical-v1

factprobe-replication-stage1-counts-canonical-v1 Occurrences of every probed entity name in the OLMo-2 PRETRAINING corpus (olmo-mix-1124, 1,117 token files, 14.10 TiB), counted as OLMo token sequences with a word boundary required at each end, both written forms kept separate. SUPERSEDES factprobe-replication-stage1-counts-olmotok-v1, which was missing each entity's canonical name: the released triples name entities by their Wikidata aliases, a field that by construction… See the full description on the dataset page: https://huggingface.co/datasets/latkes/factprobe-replication-stage1-counts-canonical-v1.

sourceHugging Facemitupdated 1mo agoView on Hugging Face
0likes21downloads
Dataset Card

factprobe-replication-stage1-counts-canonical-v1

Occurrences of every probed entity name in the OLMo-2 PRETRAINING corpus (olmo-mix-1124, 1,117 token files, 14.10 TiB), counted as OLMo token sequences with a word boundary required at each end, both written forms kept separate. SUPERSEDES factprobe-replication-stage1-counts-olmotok-v1, which was missing each entity's canonical name: the released triples name entities by their Wikidata aliases, a field that by construction excludes the canonical name, so 10,471 of 42,235 entities had no canonical name and 'Barack Obama' was never counted. It occurs 38,978,811 times. 9,476 names are new here, adding 11,576,873,076 occurrences. Verified additive: of the 141,605 names counted under both name lists, not one count changed.

Dataset Info

  • —Rows: 151081
  • —Columns: 6

Columns

ColumnTypeDescription
aliasValue('string')the name searched for, exactly as written
countleadspaceValue('int64')occurrences of the space-preceded form (' Barack Obama'), used mid-sentence
count_bareValue('int64')occurrences of the bare form, used after a line break, quote, bracket or hyphen
countValue('int64')the two summed; the forms partition occurrences
corpusValue('string')which body of training data was counted
constructValue('string')the rule that produced the count

Generation Parameters

json
{
  "experiment_name": "factprobe-replication",
  "job_id": "mll:73619",
  "cluster": "mll",
  "artifact_status": "final",
  "canary": false,
  "model": "n/a (corpus counting)",
  "script_name": "count_npy_direct.py",
  "description": "Occurrences of every probed entity name in the OLMo-2 PRETRAINING corpus (olmo-mix-1124, 1,117 token files, 14.10 TiB), counted as OLMo token sequences with a word boundary required at each end, both written forms kept separate. SUPERSEDES factprobe-replication-stage1-counts-olmotok-v1, which was missing each entity's canonical name: the released triples name entities by their Wikidata aliases, a field that by construction excludes the canonical name, so 10,471 of 42,235 entities had no canonical name and 'Barack Obama' was never counted. It occurs 38,978,811 times. 9,476 names are new here, adding 11,576,873,076 occurrences. Verified additive: of the 141,605 names counted under both name lists, not one count changed.",
  "hyperparameters": {
    "names_searched": 174729,
    "names_found": 151081,
    "token_files": 1117
  },
  "input_datasets": [
    "allenai/olmo-mix-1124",
    "Wikidata labels, fetched 2026-08-24"
  ]
}

Usage

python
from datasets import load_dataset

dataset = load_dataset("latkes/factprobe-replication-stage1-counts-canonical-v1", split="train")
print(f"Loaded {len(dataset)} rows")