Mood Journal Theme Detector

How Mood Journal Theme Detector works

Press Load model once: your browser downloads Gemma 4 E2B instruction-tuned (web build) (2.0 GB) from this site and keeps it in its cache. Then add your text and press Classify locally. The model runs on this device using your graphics card, so nothing you enter is sent to a server. Until the model is loaded, or if your device cannot run it, the tool falls back to a simple keyword check and labels the result as the fallback.

Gemma 4 E2B instruction-tuned (web build) is a small open language model from Google. The tool sends it the instruction below plus your text, asks for a JSON answer with fixed fields, and shows each field as a card. It runs at temperature 0, so the same input gives the same answer on the same device, though different graphics cards can word it slightly differently.

Inputs longer than 6,000 characters are cut to that length before they reach the model, and the result says so. The model only sees what you paste: it does not look anything up.

If what you enter mentions suicide or self-harm, the tool does not run the model. It shows crisis lines instead (in the US, call or text 988), because a trained person can help in a way a small model cannot.

The instruction the model gets

You read a mood journal entry and organise it: each feeling with when or why it came up, what seemed to help, what was hard, and the themes that run through the entry. Reflect only what the person wrote. You do not diagnose, do not name conditions or disorders, do not give medical or therapeutic advice, and never change or suggest medication doses. You only organise and reflect what the user wrote, in plain, kind language. Do not present yourself as a substitute for a therapist, doctor or other professional.

What it returns

• Feelings and moments: One entry per feeling, as 'feeling: when or why, as written'.
• What helped: Things the entry says made them feel better, quoted.
• What was hard: Things the entry says were hard or stressful, quoted.
• Themes: Short theme names that run through the entry, such as work pressure, connection or rest.

Worked example

The model's output for the built-in sample, recorded on a desktop browser. Your run may word things slightly differently.

Journal: Felt tense before client call, relieved after getting clear feedback, lonely in the evening, better after texting my sister and walking outside.

Feelings and moments

tense: before client call relieved: after getting clear feedback lonely: in the evening

What helped

getting clear feedback texting my sister walking outside

What was hard

Felt tense before client call

Themes

work pressure loneliness connection self-care

The model

The files are served from this site, versioned so a cached copy is never stale. Nothing is fetched from a third party while you use the tool.

Limitations

• Small language models make mistakes: they can miss an item, merge two, or state something more confidently than the text supports. Check the result against your input before you rely on it.
• It needs WebGPU and about 8 GB of memory, so it is offered on desktops and laptops, not on phones and tablets. Devices that cannot run it get the keyword fallback, clearly labelled.
• Results describe the input you gave, not the real-world situation behind it. Review them yourself before acting.
• This is not a medical device and does not give medical advice, a diagnosis or a treatment plan. Ask a qualified health professional about any symptom, medication or change to your routine.

Good uses

• Triage mental health and wellbeing notes before deeper review
• Prepare consistent labels for repeat workflows
• Reduce manual sorting on small batches
• Keep sensitive drafts in the browser preview