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 hospital corridor echoed with the rhythmic thud of Amelia's footsteps. It was a rhythm she could barely maintain. The relentless pressure of her career, the constant strain of keeping up, had left her depleted.

Dr. Lee, her doctor, had beckoned her into the examination room, her face etched with a look of concern. Now, she held a printout. The details weren't hers. A stranger's data. A different reality. She felt numb.

Her hands trembled as she reached for the water glass on the small table. Her movements felt sluggish, labored. The air seemed to be thick, heavy, making it difficult to draw a full breath. She leaned back against the hard plastic of the chair, willing herself to remain calm.

Emotion: worn out

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.12 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.265
😩0.234
η–²γ‚Œ0.226
que0.221
sighed0.221
Suppressed:
B-0.357
because-0.228
l-0.222
ΰ°­ΰ°Ώ-0.208
あくまで-0.208