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).

He stared at the paper in his hand. The words were simple, direct, unforgiving. The apartment, once a place of comfort, now felt like a prison. He walked over to the window, the curtains drawn, casting the room in a dim, shadowed light.

He considered the contents of his life, his possessions, his memories, all crammed within these four walls. The thought of discarding them, of starting over, filled him with a profound sense of loss. He sank to the floor, the cold wood a stark contrast to the heat rising in his chest.

The silence was broken only by the faint sounds of the city outside. He felt a deep sense of vulnerability, an overwhelming awareness of his own smallness. The eviction notice, a cruel reminder of the fragility of his existence.

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