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 flickering neon sign of the diner cast an unnatural glow on the empty street. I watched David across from me, his face etched with something beyond embarrassment. He'd been telling everyone he was a doctor, saving lives. All those late nights, all the urgent phone calls, all the heroic stories. Now I knew the truth, and the foundation of his charade seemed to crack.

I thought back to the countless times we'd laughed until we cried, the late-night talks, the secrets shared. A tightness constricted my chest. It felt wrong, somehow, to think of his deception.

I picked at my fries, the greasy smell suddenly nauseating. He looked small, insignificant. He was just a guy. And a lie had formed the base of our friendship.

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