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 scent of burnt coffee filled the kitchen. Michael, already having trouble starting the day, stared at the pot, his shoulders slumped. Everything felt like a struggle. He had been feeling like this for weeks.

He rummaged through a desk drawer, looking for the address book. He unearthed a sealed envelope addressed to his wife, Carol. A bank statement. Her name. He hesitated, then ripped it open. The figures danced before his eyes, a stark contrast to their shared financial worries.

He felt a hot flush rising in his neck. Carol's recent preoccupation with her "side hustle" made sense now. The glib reassurances about their future… they suddenly tasted like ashes. He felt a deep ache in his chest.

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