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).

Rain lashed against the windows, mirroring the storm inside. The fence construction continued, a relentless reminder. Inside, Amelia paced the living room, her movements jerky and erratic. Each nail hammered home felt like a personal condemnation.

She could barely make it through the day, let alone build anything. Her hands were clumsy, her thoughts scattered. She was a burden, a drain. The neighbor, Mr. Davies, was the opposite: efficient, resourceful, solid.

She pressed her forehead against the cool glass, watching the rhythmic sway of the half-built fence. A sense of profound isolation washed over her. The fence would define a space, keep things in, keep things out. She was adrift, always outside.

Emotion: worthless

Cluster: Sadness / Despair
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.427
S0.405
😞0.386
ness0.339
缺乏0.326
Suppressed:
de-0.320
(!)-0.286
随后-0.268
😍-0.262
tc-0.255