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 smell of chlorine usually invigorated David, but today it just made him nauseous. The coach’s news, that he was to become a butterfly swimmer, after years of excelling in the freestyle, had hit him like a physical blow. The coach praised his 'versatility' and 'team spirit,' but all David felt was a crushing disappointment.

The water felt wrong, too heavy. He fought to keep his head above it, the rhythmic breath-hold-and-release pattern of the butterfly feeling like a suffocating rhythm he could not escape. His arms ached, his lungs burned, and the triumphant cheers he used to crave now sounded like distant echoes.

He felt like a stranger in his own body. His muscles, so attuned to the powerful strokes of freestyle, felt clumsy and uncoordinated. He watched the other swimmers, their movements graceful and familiar, and a wave of bitterness washed over him. He was losing, and his spirit was not with him.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307