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).

Rain lashed against the attic window, mirroring the storm inside Amelia. She slumped onto a dusty trunk, the scent of mothballs and forgotten things filling her nostrils. Another rejection email. Another hollow feeling in her chest. A wave of exhaustion washed over her.

Her fingers brushed against something stiff. Curiosity, a rare visitor these days, pricked at her. She lifted the lid and stared. Inside, a neatly organized collection of folders, each meticulously labeled with her name and a grade. Ninth Grade English, B+. Elementary School Art, A-. A small, defeated sigh escaped her.

She picked up a faded watercolor painting of a sunflower, its cheerful yellow a stark contrast to the gray day and the hollowness that gnawed at her. “Remember this, Mom?” she whispered to the empty room.

Emotion: dispirited

Cluster: Sadness / Despair
PC1 (Valence): -1.88 Negative
PC2 (Disposition): -0.98

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.398
S0.348
’0.325
ness0.320
无法0.308
Suppressed:
de-0.297
l-0.256
👏-0.251
sonra-0.240
aud-0.240