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 gentle patter of rain against the window, the warmth of the fireplace, the aroma of freshly baked bread – Michael was submerged in complete relaxation. He was reading, and life seemed simple. A notification on his phone, a routine message from his doctor's office, disrupted the bliss. He opened the file, his fingers already anticipating the mundane details. The information was not his. Not even close.

A jolt of pure shock ran through him, freezing him in place. The warmth of the fire no longer reached him; he felt a sudden chill. The bread's aroma became sickening. He read the file again, and again. It was a mistake. A devastating one.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350