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 found himself tapping his pen incessantly against the desk, a rapid staccato against the otherwise quiet hum of the server room. The chill that had settled in his bones seemed to originate from the pit of his stomach. A tight knot of tension was coiled there. Mark had always played by the rules, worked diligently, and kept his head down. Now, hearing that Liam, the new intern, was pulling in more than him, a wave of something bitter and acrid swept through him. He glanced at the overflowing inbox, a task list that never seemed to shrink. He wanted to shout, to break something. Instead, he took a deep, shaky breath and forced a smile when Liam walked past.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199