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 bustling city blurred around Michael as he wandered, each step feeling heavier than the last. He had lost his job. He felt like a discarded toy, forgotten and unloved. He felt a profound sense of isolation.

He bumped into a figure. It was Susan, who had tormented him endlessly during their school years. He recoiled instinctively, expecting scorn. Instead, she had a different look on her face.

"Michael? Wow. Fancy seeing you here," Susan said. "I’m a therapist, actually. Are you alright? You look… troubled." He swallowed hard, his throat constricting. The streets seemed to spin.

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