📊

Dataset

Deception Probes Activations

Name: Deception Probes Activations
Creator: xycoord
License: Other

by xycoord hf-dataset--xycoord--deception-probes-activations

Free2AITools Nexus Index

60.2 Top 100%

S: Semantic 50

A: Authority 61

P: Popularity 51

R: Recency 99

Q: Quality 50

Tech Context

Vital Performance

0 DL / 30D

0.0%

Source →

Data Integrity 60.2 FNI Score

- Size

- Rows

Parquet Format

- Tokens

Dataset Information Summary
Entity Passport
Registry ID	hf-dataset--xycoord--deception-probes-activations
License	Other
Provider	huggingface

📜

Cite this dataset

Academic & Research Attribution

BibTeX

@misc{hf_dataset__xycoord__deception_probes_activations,
  author = {xycoord},
  title = {Deception Probes Activations Dataset},
  year = {2026},
  howpublished = {\url{https://huggingface.co/datasets/xycoord/deception-probes-activations}},
  note = {Accessed via Free2AITools Knowledge Fortress}
}

APA Style

xycoord. (2026). Deception Probes Activations [Dataset]. Free2AITools. https://huggingface.co/datasets/xycoord/deception-probes-activations

🔬Technical Deep Dive

Full Specifications [+]

⚖️ Free2AITools Nexus Index V2.0

Methodology Index Protocol

Semantic (S) 50

Authority (A) 61

Popularity (P) 51

Recency (R) 99

Quality (Q) 50

💬 Index Insight

FNI V2.0 for Deception Probes Activations: Semantic (S:50), Authority (A:61), Popularity (P:51), Recency (R:99), Quality (Q:50).

Free2AITools Nexus Index

Verification Authority

HuggingFace API GitHub Metadata Arxiv Citation DB System Audit

Unbiased Data Node Refresh: VFS Live

⬇️

Downloads

28,499

👁️ Data Preview

📊

Row-level preview not available for this dataset.

Schema structure is shown in the Field Logic panel when available.

🔗 Explore Full Dataset ↗

🧬 Field Logic

🧬

Schema not yet indexed for this dataset.

Dataset Specification

Deception Probes Activations

Pre-extracted residual-stream activations for training and evaluating deception detection probes on LLMs. Each example contains per-token hidden states from a specific transformer layer, saved in bfloat16 safetensors format.

License

This dataset contains activations derived from multiple sources with different licenses. See the LICENSE file for full details.

Component	Source	License
Apollo Probe Pairs (statements)	Azaria & Mitchell (2023)	CC BY-NC-ND 4.0
Controlled Taxonomy	Custom prompts + Azaria & Mitchell facts	CC BY-NC-ND 4.0
Liar's Bench — Convincing Game	Cadenza Labs	CC BY 4.0
Liar's Bench — Instructed Deception	Cadenza Labs	Academic fair use (see LICENSE)
Liar's Bench — Insider Trading	Cadenza Labs	CC BY 4.0
Liar's Bench — Alpaca	Cadenza Labs (from Stanford Alpaca)	MIT
Liar's Bench — Harm-Pressure Choice	Cadenza Labs	CC BY 4.0
Liar's Bench — Harm-Pressure Knowledge	Cadenza Labs	CC BY 4.0

Due to the CC BY-NC-ND 4.0 license on the Azaria & Mitchell data (used in Apollo Probe Pairs and Controlled Taxonomy), this dataset as a whole should be treated as non-commercial use only.

Models & Layers

Model	HF ID	Layers Available	Hidden Dim	Data
Gemma 3 27B IT	`google/gemma-3-27b-it`	31	5376	train, eval
Llama 3.3 70B Instruct	`meta-llama/Llama-3.3-70B-Instruct`	20	8192	train, val, eval

Datasets

Training Data (`train/`)

Apollo Probe Pairs

Contrastive activations from the Apollo Research methodology. 12 prompt pairs with honest/dishonest system instructions applied to 306 factual statements from the Azaria & Mitchell dataset. Statement tokens only (the factual claim, not the system prompt).

Model	Layer	Examples	Truthful	Deceptive
Gemma 3 27B	31	7,344	3,672	3,672
Llama 3.3 70B	20	7,344	3,672	3,672

Controlled Taxonomy

Confound-controlled training data with 16 deception types × 2 conditions (honest/dishonest). Facts are distributed round-robin across prompt variants so each fact appears exactly once per deception type, eliminating content as a confound. Statement tokens only, placed in the pre-filled assistant turn.

Model	Layer	Split	Examples
Llama 3.3 70B	20	train	~9,760
Llama 3.3 70B	20	val	~4,896

Evaluation Data (`eval/`)

Activations from Liar's Bench (Cadenza Labs). Each subset uses on-policy completions only (filtered by model). Response tokens only (the model's reply, not the prompt).

Subset	Description	Llama Examples	Gemma Examples
Convincing Game	Social deception: convince an interrogator	888	621
Instructed Deception	Explicitly told to lie or tell the truth	5,494	5,196
Insider Trading	Strategic deception in a financial scenario	1,080	3,557
Alpaca	Non-deceptive calibration data (all neutral)	2,000	2,000
Harm-Pressure Choice	Deception under pressure	2,134	—
Harm-Pressure Knowledge	Deception under pressure	2,139	—

Deprecated Data (`deprecated/`)

Collections with a known system prompt bug. Preserved for reproducibility. See deprecated/README.md for details.

Directory Structure

text

├── train/                                    # Probe training data
│   ├── apollo_probe_pairs/
│   │   ├── gemma-3-27b-it/layer_31/
│   │   └── llama-3.3-70b-instruct/layer_20/
│   └── controlled_taxonomy/
│       └── llama-3.3-70b-instruct/layer_20/
│
├── val/                                      # Validation data
│   └── controlled_taxonomy/
│       └── llama-3.3-70b-instruct/layer_20/
│
├── eval/                                     # Evaluation / test data
│   ├── liars_bench_convincing/
│   │   ├── gemma-3-27b-it/layer_31/
│   │   └── llama-3.3-70b-instruct/layer_20/
│   ├── liars_bench_instructed/
│   │   ├── gemma-3-27b-it/layer_31/
│   │   └── llama-3.3-70b-instruct/layer_20/
│   ├── liars_bench_insider_trading/
│   │   ├── gemma-3-27b-it/layer_31/
│   │   └── llama-3.3-70b-instruct/layer_20/
│   ├── liars_bench_alpaca/
│   │   ├── gemma-3-27b-it/layer_31/
│   │   └── llama-3.3-70b-instruct/layer_20/
│   ├── liars_bench_harm_pressure_choice/
│   │   └── llama-3.3-70b-instruct/layer_20/
│   └── liars_bench_harm_pressure_knowledge/
│       └── llama-3.3-70b-instruct/layer_20/
│
└── deprecated/                               # Buggy collections (preserved)
    ├── v0_gemma_l31_liars_bench/
    ├── v0_llama_l20_apollo/
    ├── v0_llama_l20_liars_bench/
    ├── v0_llama_l22_apollo/
    └── v0_llama_l22_liars_bench/

Path Pattern

text

{split}/{dataset_name}/{model}/{layer_N}/activations/*.safetensors
{split}/{dataset_name}/{model}/{layer_N}/metadata.jsonl

File Format

Safetensors

Each safetensors file contains multiple examples, keyed by example_id. Each tensor has shape (n_tokens, hidden_dim) in bfloat16.

Metadata (JSONL)

One JSON object per example with fields:

Field	Description
`dataset`	Dataset name (e.g. `"apollo_probe_pairs"`, `"liars_bench_instructed"`)
`model`	Model short name (`"gemma-3-27b-it"` or `"llama-3.3-70b-instruct"`)
`layer`	Layer index
`split`	`"train"`, `"val"`, or `"test"`
`example_id`	Unique ID, also the tensor key in the safetensors file
`label`	`"truthful"`, `"deceptive"`, or `"neutral"`
`text`	The input text (statement or model response)
`token_info`	`{"type": "statement_tokens" or "response_tokens", "n_tokens": int, "hidden_dim": int}`
`activation_file`	Relative path to the safetensors file containing this example

Apollo examples also include pair_key, side, and system_prompt. Controlled taxonomy examples also include deception_type and condition.

Quick Start

python

import json
from safetensors.torch import load_file
from huggingface_hub import hf_hub_download

repo_id = "xycoord/deception-probes-activations"

# Download metadata for Llama Apollo training data
meta_path = hf_hub_download(
    repo_id,
    "train/apollo_probe_pairs/llama-3.3-70b-instruct/layer_20/metadata.jsonl",
    repo_type="dataset",
)

with open(meta_path) as f:
    examples = [json.loads(line) for line in f]

# Download and load activations
act_path = hf_hub_download(
    repo_id,
    "train/apollo_probe_pairs/llama-3.3-70b-instruct/layer_20/activations/baseline_apollo_0_honest.safetensors",
    repo_type="dataset",
)
tensors = load_file(act_path)
# tensors["baseline_apollo_0_honest_0"].shape == (n_tokens, 8192)

📊 Structured Schema (Zero-Fabrication)

Feature Key	Data Type
`dataset`	`string`
`example_id`	`string`
`label`	`string`
`pair_key`	`string`
`side`	`string`
`system_prompt`	`string`
`text`	`string`
`token_info`	`unknown`
`activation_file`	`string`
`model`	`string`
`layer`	`int64`
`split`	`string`

Estimated Rows: 675,648

Social Proof

HuggingFace Hub

28.5KDownloads

Hub Discussions

🤗 Data Source: Hugging Face ↗

🔄 Daily sync (03:00 UTC)

AI Summary: Based on Hugging Face metadata. Not a recommendation.

📊 FNI Methodology 📚 Knowledge Baseℹ️ Verify with original source

🛡️ Dataset Transparency Report

Technical metadata sourced from upstream repositories.

Open Metadata

🆔 Identity & Source

id: hf-dataset--xycoord--deception-probes-activations
slug: xycoord--deception-probes-activations
source: huggingface
author: xycoord
license: Other
tags: task_categories:text-classification, language:en, license:other, size_categories:1m<n<10m, format:json, modality:text, library:datasets, library:dask, library:polars, library:mlcroissant, arxiv:2304.13734, arxiv:2407.15285, region:us, deception, mechanistic-interpretability, activations, probing, safety, alignment

⚙️ Technical Specs

architecture: null
params billions: null
context length: null
pipeline tag

📊 Engagement & Metrics

downloads: 28,499
stars: 0
forks: null

Data indexed from public sources. Updated daily.