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 empty gallery echoed his failure. Years dedicated to art, and here he was, staring at his unsold canvases. A hollow laugh escaped him. He felt like the shadows themselves.

His sister, knowing his mood, had invited him over for dinner. His nephew, Leo, ran to him, brandishing a crayon drawing. It was a person made of sunlight.

โ€œThis is Solara,โ€ Leo announced. โ€œShe shows me how to be brave.โ€

Daniel examined the drawing, a strange tightness in his throat. Solara. The sun-person who had guided him through all his childhood fears. He realized he hadn't created anything worthwhile. He had failed to inspire.

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