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 coffee tasted like ash in Clara’s mouth. She sat at her kitchen table, staring at the letter she’d found tucked away in a box of old photos. Her hand was shaking as she read it, the words penned with a frantic energy she no longer possessed. It was a plea to her sister, an apology for a childhood slight.

Her head throbbed. She leaned forward, resting her forehead on the cool surface of the table. A dullness had settled over her, a blanket of quiet despondency that had been present all day. The bright morning was gone, leaving the afternoon empty.

She felt a strange disconnect, as if she were watching her life from a great distance. The sharp sting of regret, once so potent, had mellowed into a pervasive numbness. She placed the letter gently back in its envelope, the gesture feeling as heavy as the stone that seemed to have materialized in her heart.

Emotion: droopy

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.15 Negative
PC2 (Disposition): -1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.303
无法0.252
いつも0.238
😔0.230
:(0.228
Suppressed:
l-0.360
P-0.356
B-0.339
de-0.334
!-0.239