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 office was quiet, save for the rhythmic click of the keyboard and the distant hum of the ventilation system. The air tasted stale, heavy. He watched the young man intently, his brow furrowed in concentration. He felt a sudden, sharp pang.

He rubbed the back of his neck, trying to ease the tension that had taken residence there. He felt a hollowness, an emptiness. He remembered the camaraderie, the shared goals, the late nights fueled by caffeine and ambition. That was all in the past now.

“Okay, now let’s review the security protocols,” he said, his voice a low monotone. He looked at the window, the bright, bustling street scene outside. He felt a sense of… completion, maybe. Or was it just exhaustion? He wasn’t sure anymore.

Emotion: reflective

Cluster: Nostalgia / Reflection
PC1 (Valence): -1.27 Negative
PC2 (Disposition): -1.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.650
L0.547
如今0.402
H0.353
own0.319
Suppressed:
la-0.687
de-0.659
l-0.388
/-0.324
--0.320