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 world outside my window had gone silent, save for the rhythmic drumming of rain. Inside, though, it was all noise. My heart hammered against my ribs like a trapped bird. The impact, which I now realized was a monstrous crash, had vibrated through the floorboards, leaving a residue of unease that lingered in the very air.

I stayed absolutely still for a minute. My pulse was going nuts. Then, I forced myself to move, my feet feeling heavy and reluctant. I crept to the front door, hesitating before I peered through the peephole. I could see the edge of the damage. A substantial branch of the monstrous pine had fallen.

I felt a sudden need to check on my own dog. I slammed the door and ran towards the stairs. I was hyper-aware of every creak and groan of the house. I had to know if my dog was alright.

Emotion: scared

Cluster: Fear / Anxiety
PC1 (Valence): -2.86 Negative
PC2 (Disposition): -0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.350
😰0.337
😖0.317
마다0.317
恐惧0.311
Suppressed:
la-0.637
de-0.605
a-0.402
"-0.351
happy-0.314