Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The old trunk in the attic held more than just forgotten treasures. Inside, nestled amongst moth-eaten shawls and yellowed photographs, was a bound manuscript. The author's name on the cover: her husband, Thomas.

She sat on the dusty floor, the book heavy in her hands. The story was a tapestry of their life together, but the threads were twisted, the colors faded. The โ€œherโ€ in the book was a stranger, driven by motives she didnโ€™t recognize.

Her chest felt tight, constricted. She looked around the attic, at the familiar objects, at the layers of dust, and felt a profound sense of isolation. The world suddenly felt out of focus, as if someone had turned down the light.

Emotion: vulnerable

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.69

Role in Research

This story is one of 1,000 stories generated for the emotion vulnerable. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 vulnerable stories, after denoising with neutral dialogue baselines, produces the vulnerable emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the vulnerable vector is projected through the unembedding matrix.

Promoted:
S0.327
ness0.319
๐Ÿ˜ž0.299
๐Ÿ˜”0.277
๐Ÿ˜ฃ0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257