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 old woman's collection of porcelain dolls, arranged on the shelves, suddenly felt menacing. Emily hated dusting, but she had to. The sunlight slanting through the windows seemed to expose every flaw in the house, every imperfection. She'd been feeling this way for a while, just… off. The thought of starting any project seemed overwhelming.

A phone call, the voice of the officer, cold and clinical. He delivered the facts. The house's secret was laid bare. Emily stared at the dolls, their vacant eyes staring back at her. The house, her haven, was now a cage, filled with a history she couldn't escape. The urge to flee became almost unbearable.

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
πŸ‘-0.251
sonra-0.240
aud-0.240