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 small apartment felt cold, even with the heater cranked up. She paced back and forth, the worn carpet offering little comfort. The rejection, a brick wall in her path, felt like something she couldn’t scale.

She flicked through the channels on the television, finding nothing that could hold her attention. The world outside seemed distant, irrelevant. Every advertisement, every sitcom, felt like a deliberate reminder of the life she wouldn't have.

She picked up her phone, then put it back down, unable to bring herself to call anyone. There was no one who could understand the particular hollowness of this situation. The silence of the apartment was suffocating, echoing her own internal emptiness.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
’0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🤩-0.225
que-0.216