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 stale air in the apartment pressed in on her. The world felt muffled, distant. A knock at the door. Her neighbor, Mr. Garcia, stood there, his face flushed with the energy she couldn't muster. A watering can swung gently at his side.

"Good afternoon! Beautiful day, right? Listen… I was thinking… the best light is in your backyard! I was hoping… I was hoping I could put in a garden, and…” he stammered a little.

She closed her eyes briefly, the brightness outside an assault. The thought of the effort of anything, even a simple response, was overwhelming.

"Fine," she mumbled, not bothering to meet his eyes. The interior of her mind was a wasteland; his garden idea was just another thing.

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