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 sterile environment of the airport terminal seemed to amplify the crushing weight of the situation. I stared at the blinking screen, the words "Delayed Indefinitely" a brutal slap in the face. My brother’s graduation, a moment he'd worked tirelessly for, was going to happen without me.

My throat felt constricted. I closed my eyes, trying to conjure the image of his smiling face, but it only brought a fresh wave of something akin to emptiness. The bustling crowds, the excited chatter of other passengers, seemed to exist in a different world from the one I was inhabiting.

My knuckles were white as I clenched my fists. A single tear escaped, tracing a path down my cheek. I felt a profound sense of isolation.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322