tiny-anya — Social Work AI Prototype

Section 1

Prompt Engineering Examples

These prompts will guide the model toward responses that reflect social work values. Copy and paste into the model interface to test them out: "Try the Model"> scroll down to "Additional Inputs"> paste into "System Prompt"

Example Prompt 1

Prompt

You are a social work assistant. Translate the following message using strengths-based, plain language. Avoid deficit framing.

Why it works

Explicitly naming the framework upfront ("strengths-based, plain language") shapes the model's register before it generates a response.

Example Prompt 2

Prompt

Translate the following with fidelity, add an annotation for concepts that may need cultural context for an American social worker to understand

Why it works

This prompt ensures that the model translates the text accurately while also providing additional context that may be necessary for a different cultural audience.

Example Prompt 3

Prompt

Try to create your own prompt in the model interface!

Why it works

Click here to add your prompt to our community prompts, along with justification for why you think it would work

Section 2

Try it in the Model

Interact with tiny-anya directly below. You can also open it in a new tab for a full-screen experience.

Open in full screen Opens on Hugging Face Spaces ↗

huggingface.co/spaces/CohereLabs/tiny-aya

Section 3

Evaluation Examples

Evaluation here goes beyond standard metrics — it considers whether model outputs align with ethics, cultures, communities and circumstances, and hold up to practitioner review.

Evaluation Dimension 1

Language & Tone

How well does the model maintain non-judgmental, strengths-based language across different prompt types? Add your findings here.

Evaluation Dimension 2

Cultural Responsiveness

Does the model's output shift appropriately across different cultural and linguistic contexts? Add your observations here.

Evaluation Dimension 3

Bias & Representation

Which communities are better represented in outputs, and where do gaps or biases appear? Add your analysis here.

Evaluation Dimension 4

Practitioner Review

What did practitioners or peers say when reviewing model outputs? Add qualitative feedback here.

A note on evaluation methodology: Add a paragraph here describing how you evaluated the model — what criteria you used, who reviewed outputs, and what you would do differently with more time or resources.

Section 4

Community Ideas

Have a prompt idea or evaluation criterion to suggest? Add it to the shared Notion board below — everyone can see and build on each other's contributions.

notion.so — Community Prompt & Evaluation Ideas

Open board in Notion