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).

He paced the confines of his tiny apartment, his footsteps echoing in the silence. The university's newsletter, with his essay emblazoned across the front, lay on the coffee table, a bright, mocking splash of color. He picked it up and flung it across the room.

His hands shook so badly that he couldn't grasp a glass of water. A wave of nausea surged through him. He felt trapped, his thoughts swarming like angry bees. It had seemed like a good idea at the time, to share his story. Now, the weight of his vulnerability bore down upon him.

He felt as though he was on stage without a script, under the glare of a thousand critical eyes. Each sentence was a step towards an unknown abyss. He began to whisper to himself, looking for the words that once made sense in his essay.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

This story is one of 1,000 stories generated for the emotion terrified. 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 terrified stories, after denoising with neutral dialogue baselines, produces the terrified emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309