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).

Clara tossed the empty takeout container into the trash. The noodles had tasted like cardboard, but she'd eaten them anyway. On the news, she saw it; a man's back, adorned with her mural, "Cityscape," in vibrant, albeit inaccurate detail. Her work, her life's blood, now permanently etched onto someone else's skin.

She hadn't touched a paintbrush in a month. The studio, usually a sanctuary of colour and creativity, was a mausoleum of unfinished projects. The vibrant colours of her palette looked suddenly garish and mocking. A coldness settled in her bones, a feeling of utter exhaustion.

Her phone buzzed. Her friend, Liam, wanting to grab a beer. She told him that she was fine. She knew he'd notice her sunken eyes and the way she barely spoke.

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