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 bookstore felt oppressively bright, the scent of fresh ink and paper a mocking reminder. Daniel saw his name, bold and stark, on the cover: β€œThe Quiet Life of Daniel Harding.” He’d always been a creature of shadows, a collector of small moments. Now, those moments were being hawked.

He ran a hand through his hair, the gesture frantic. His stomach twisted, an uncomfortable knot forming. He wanted to disappear, to become invisible. He felt a desperate urge to flee, to escape this public exposure. He felt a great ache in his core, which spread outwards.

He grabbed a copy, the cover searing his fingertips. He felt an intense desire to tear it to shreds.

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
πŸ˜”0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322