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 house was cold, the air thick with an oppressive stillness. He moved through his days like a ghost, a shadow of the man he used to be. The sun seemed to have forgotten how to shine. His sister had sent a new drawing from her son, Ethan. It depicted a whimsical, multi-limbed creature with a perpetually surprised expression. He recognized it instantly. Pip, his childhood best friend.

He felt a pull, a longing he couldn't name. Pip, with all his odd limbs and weird quirks, was a friend like no other. He clutched the drawing to his chest, the paper crinkling under his fingers. A tiny spark of something, a forgotten feeling, flickered within him.

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