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 air in Mr. Ramirez’s apartment felt stale, thick with unspoken thoughts. He spent most of his time in his armchair, staring at the television, the noise a dull background to the silence within him. Getting dressed for work felt like a monumental task. The energy to get through the day felt like it was gone.

A package arrived, addressed in a familiar, sprawling handwriting. Inside, wrapped in colorful tissue paper, was a collection of his favorite books, a well-loved copy of *One Hundred Years of Solitude* on top. A small, wooden bookmark, carved with a simple design, rested within the pages. A note was tucked inside: "Thanks for sharing your stories, Mr. Ramirez. - David.” He breathed in deeply, the scent of the old books bringing a flicker of comfort.

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