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 of the greasy spoon diner. Amelia hunched over a lukewarm cup of coffee, the steam doing little to warm her chilled hands. The radio crackled, then a familiar chord progression filled the air. Her song. Or, well, *her* song, now being sung by a saccharine-voiced pop starlet. A wave of nausea churned in her stomach. She felt her chest tighten, as if an invisible weight was pressing down. Sheโ€™d poured her heart into that song, every syllable a fragile piece of herself. Now, it was a hollow shell, mass-produced and stripped of its soul.

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
๐Ÿ˜ž0.386
ness0.339
็ผบไน0.326
Suppressed:
de-0.320
(!)-0.286
้šๅŽ-0.268
๐Ÿ˜-0.262
tc-0.255