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 late afternoon light cast long shadows across the park. Leo sat on a bench, watching children play, their laughter a discordant note in the quiet of his interior. The sketches he had brought with him lay untouched in his bag. He couldn't bring himself to create.

Then, a cyclist whizzed past, sleeves pulled up to reveal a familiar image. A direct replica of his "Guardian Angel" design, tattooed on the calf. It was exquisite, undeniably beautiful, and yet, it was a blow.

He felt the familiar tightening in his chest. A sudden, almost physical sensation of being diminished. He pushed himself off the bench, the world tilted slightly beneath his feet. The park, once a source of inspiration, seemed to mock him.

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