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 stared at the blank screen of his laptop, the cursor blinking accusingly. The project felt monumental, an insurmountable mountain of research and writing. A knot formed in his stomach. He rubbed his temples, searching for something, anything, to break the monotony.

His gaze landed on a box in the basement, a legacy of his mother’s recent move. The label read, "Mom's Memories". With a sigh, he descended the stairs. He’d seen this box countless times before, never gave it a second thought.

Inside, amidst old photo albums and trinkets, he discovered a collection of his mother's creative writing assignments. A short story about a talking cat, a poem about falling leaves. An unexpected spark of curiosity ignited in his chest. He forgot the task ahead, as he read through his mother's work.

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