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).

“I don’t understand,” Maya had muttered, repeating the phrase to the hospital’s patient liaison over and over again. The explanation, a convoluted account of data entry errors, provided no comfort. The news of the medical record mix-up had left her feeling adrift, cast away.

She found herself staring at the ceiling of her bedroom, the intricate patterns of the paint blurred by unshed tears. The familiar comfort of her surroundings felt tainted, the very air thick with a sense of violation. This wasn’t just a breach of privacy, it was an intrusion, a violation.

The dinner she had prepared, a favorite recipe, sat cooling on the stove, untouched. She hadn't eaten a single bite. Maya felt like her very body was betraying her, mirroring her emotional turmoil in a lack of appetite. She hugged herself, trying to ward off the chill that had settled deep within her bones.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259