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 kept seeing the glint of gold, even in the dim library lighting. Michael couldn't shake the image of his sister, Clara, her face lit with the triumphant joy of inheriting the family estate. He, on the other hand, had received a small, leather-bound journal. Its pages, filled with illegible script, were meant to chronicle a lineage of alchemists.

His stomach churned, a feeling that intensified with each clumsy translation attempt. He felt a creeping dread, the way the air pressed down on his lungs, making it difficult to breathe.

He slammed the journal shut. A small, dry leaf detached from the spine and fell onto the desk.

"Great," he muttered, picking up the leaf and crushing it into a fine dust.

Emotion: disturbed

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.318
ness0.309
无法0.300
😞0.297
worse0.284
Suppressed:
de-0.829
la-0.666
a-0.392
B-0.301
"-0.290