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 opulent surroundings of the ballroom felt like a cage. Finn watched his sister, Esme, accept congratulations from the assembled guests, her laughter echoing through the room. He felt small, invisible.

The solicitor announced the distribution of their grandfather’s estate. Esme inherited the family's vast financial holdings, a legacy of wealth and power. Finn received a battered old motorcycle and a collection of his grandfather's tools.

He found himself outside, the cool night air doing nothing to clear his head. He kicked at a pebble, his anger and disappointment building. The motorcycle, now his only possession, felt like a burden, not a freedom. He didn't know how to ride it, or how to fix it.

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