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 vibrant energy of the art fair once invigorated her, but now it left her exhausted. The forced smiles, the endless conversations, the pressure to sell – it was all too much. She’d withdrawn, her booth a lonely island in a sea of eager artists. She just stood there, pretending to be engaged, though she wasn't.

She saw it during a phone call with her sister. Her sister sent her a photo. A girl on the phone had a tattoo of one of her paintings. It was so strange to see it on someone else.

Her voice cracked. "That's… that's one of mine."

Her sister's laughter felt like a knife. She immediately ended the call, and shut off her phone. She looked at her own hands, then began to cry.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307