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).

It was addressed to "Mr. Harold Finch," someone I'd never met. The name, however, triggered a familiar ache in my chest. I'd lived in this flat for a long time. I hadn’t given much thought to the previous occupants.

I stood there for a moment, the package heavy in my hands. The edges of the cardboard felt rough against my skin. I went inside and set the box on the kitchen table.

Inside, I discovered a collection of vintage records. I pulled one out, a vinyl disc, and held it to my nose, inhaling the scent of dust and time. I had a record player in the living room. I felt a yearning, a desire to hear the music.

Emotion: sentimental

Cluster: Nostalgia / Reflection
PC1 (Valence): 0.89 Positive
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.677
fondly0.567
memories0.503
曾0.488
reminis0.466
Suppressed:
either-0.515
or-0.488
[]-0.473
--0.408
не-0.396