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 diner buzzed around Amelia. Laughter, clinking cutlery, the rhythmic whoosh of the coffee machine - a symphony of sounds she used to find comforting. Now, it just grated. She sat alone in a booth, picking at her eggs. Her phone vibrated, a notification, but she ignored it. A letter had arrived that morning, sent by a person she had not thought of in ages.

It was an apology for a callous act. A teenage drama, a forgotten transgression. She pushed the plate of food away, appetite vanished. The words, so carefully chosen, felt clumsy and inadequate. She felt the world closing in, each breath a struggle. The bright lights of the diner seemed to dim, reflecting the emptiness she felt.

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