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 air in the museum felt thick and heavy to Thomas. His intricately woven tapestries, once hidden in his grandmother's attic, were now the centerpiece of a prestigious exhibit, stolen and recontextualized without his involvement. He felt an intense sting.

A guide, with a practiced smile, pointed out the "unique techniques" employed in the pieces, completely missing the emotional undercurrent, the personal history woven into each thread. Thomas felt a burning sensation in his throat, a desire to explain, to correct, to reclaim.

He leaned against a cold marble pillar, a wave of nausea washing over him. The grand space, usually a place of inspiration, now felt alien and suffocating.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322