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 windowpane, mirroring the dreary emptiness that had settled in Sarah's soul. She sighed, pushing away from her desk, the half-finished crossword puzzle mocking her. The hum of the refrigerator seemed to amplify the silence. A notification popped up on her phone: a link to an article, shared by a friend, with a catchy title. Curiosity, a rare visitor these days, pricked at her.

She clicked the link, her heart rate quickening with a flicker of unease. It was a blog post, a compilation of embarrassing teenage diary entries. Scrolling through the comments, she recognized her name. With a surge of dread, she clicked on the link.

There it was. Her diary. The one she thought was safely tucked away, a relic of a lifetime ago. A wave of heat washed over her face. She felt a tingling sensation in her fingertips.

Emotion: bored

Cluster: Fatigue / Lethargy
PC1 (Valence): 0.04 Positive
PC2 (Disposition): -1.11

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.387
’0.350
unremarkable0.325
दिलचस्प0.323
seemingly0.319
Suppressed:
B-0.592
P-0.304
because-0.292
因为它-0.287
ancar-0.278