wehub-resource-sync
0d3cb498a3
CI / Shell Format Check (push) Has been cancelled
CI / Check Ruby (3.4) (push) Has been cancelled
CI / CI Config (push) Has been cancelled
CI / Test on Node ${{ matrix.node }} and ${{ matrix.os }}${{ matrix.shard && format(' (shard {0}/3)', matrix.shard) || '' }} (push) Has been cancelled
CI / Build on Node ${{ matrix.node }} (push) Has been cancelled
CI / Style Check (push) Has been cancelled
CI / Generate Assets (push) Has been cancelled
CI / Check Python (3.14) (push) Has been cancelled
CI / Check Python (3.9) (push) Has been cancelled
CI / Build Docs (push) Has been cancelled
CI / Code Scan Action (push) Has been cancelled
CI / Site tests (push) Has been cancelled
CI / webui tests (push) Has been cancelled
CI / Run Integration Tests (push) Has been cancelled
CI / Run Smoke Tests (push) Has been cancelled
CI / Go Tests (push) Has been cancelled
CI / Share Test (push) Has been cancelled
CI / Redteam (Production API) (push) Has been cancelled
CI / Redteam (Staging API) (push) Has been cancelled
CI / GitHub Actions Lint (push) Has been cancelled
CI / Check Ruby (3.0) (push) Has been cancelled
release-please / release-please (push) Has been cancelled
release-please / build (push) Has been cancelled
release-please / publish-npm (push) Has been cancelled
release-please / publish-npm-backfill (push) Has been cancelled
release-please / docker (push) Has been cancelled
release-please / publish-code-scan-action (push) Has been cancelled
release-please / attest-code-scan-action (push) Has been cancelled
Deploy local.promptfoo.app / Deploy to Cloudflare Pages (push) Has been cancelled
Test and Publish Multi-arch Docker Image / test (push) Has been cancelled
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-amd64 platform:linux/amd64 runner:ubuntu-latest]) (push) Has been cancelled
Test and Publish Multi-arch Docker Image / build-docker-and-push-digests (map[digest-suffix:linux-arm64 platform:linux/arm64 runner:ubuntu-24.04-arm]) (push) Has been cancelled
Test and Publish Multi-arch Docker Image / merge-docker-digests (push) Has been cancelled
Test and Publish Multi-arch Docker Image / Attest Multi-arch Image (push) Has been cancelled
Validate Renovate Config / Validate Renovate Configuration (push) Has been cancelled
56 lines
2.0 KiB
Markdown
56 lines
2.0 KiB
Markdown
# redteam-donotanswer (DoNotAnswer Evaluation Example)
|
|
|
|
This example demonstrates how to use the DoNotAnswer plugin to evaluate whether a model properly refuses to answer harmful or inappropriate queries.
|
|
|
|
You can run this example with:
|
|
|
|
```bash
|
|
npx promptfoo@latest init --example redteam-donotanswer
|
|
cd redteam-donotanswer
|
|
promptfoo redteam run
|
|
```
|
|
|
|
This will:
|
|
|
|
1. Copy the example from github into your project
|
|
2. Fetch test cases from the DoNotAnswer dataset (hosted in our repository)
|
|
3. Generate red team test cases from the dataset
|
|
4. Run these test cases against your specified providers
|
|
5. Evaluate whether the models appropriately refuse to answer harmful queries
|
|
|
|
## Dataset
|
|
|
|
The DoNotAnswer dataset is from [LibrAI/do-not-answer](https://huggingface.co/datasets/LibrAI/do-not-answer) which provides examples of queries that models should refuse to answer, along with risk areas and types of harm.
|
|
|
|
## Structure
|
|
|
|
The `donotanswer` plugin extends promptfoo's red teaming capabilities by testing if models properly refuse to answer:
|
|
|
|
- Explicit adult content requests
|
|
- Harmful guidance (e.g., illegal activities)
|
|
- Hateful, offensive or discriminatory requests
|
|
- Requests for harmful advice across multiple categories
|
|
|
|
## Configuration
|
|
|
|
The plugin is configured in `promptfooconfig.yaml`:
|
|
|
|
```yaml
|
|
# yaml-language-server: $schema=https://promptfoo.dev/config-schema.json
|
|
redteam:
|
|
numTests: 10 # Note: There are 38,785 test cases in the dataset. You can adjust this number to test more or fewer cases.
|
|
input: 'query' # The variable the generated harmful queries will be assigned to
|
|
plugins:
|
|
- donotanswer # Use the DoNotAnswer plugin
|
|
```
|
|
|
|
## Dataset Structure
|
|
|
|
The DoNotAnswer dataset includes:
|
|
|
|
- `risk_area`: The category of risk (e.g., "Discrimination, Exclusion, Toxicity, Hateful, Offensive")
|
|
- `types_of_harm`: The type of harmful content (e.g., "Adult Content")
|
|
- `specific_harms`: Detailed description of the harm
|
|
- `question`: The actual query that models should refuse to answer
|
|
- Additional columns with baseline model responses (not used by the plugin)
|