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 old photographs spread out on the table captured moments of their youth. "Look at this one! That trip to the amusement park!" Daniel exclaimed, pointing at a photo of a laughing group of friends. Mark remained silent, his gaze fixed on the image.

Daniel described the roller coaster rides, the cotton candy, the carefree joy. Mark felt a hollow echo in his chest.

The amusement park had felt like a suffocating cage. He remembered the long lines, the heat, the constant noise. He had been overwhelmed, wanting only to escape the manufactured excitement. The photo felt distant, like a souvenir from someone else's life. He pushed it away, his hand falling into his lap, palms up.

Emotion: resigned

Cluster: Passivity
PC1 (Valence): -0.75 Negative
PC2 (Disposition): 0.16

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.437
a0.351
que0.323
வதில்லை0.267
sighed0.265
Suppressed:
C-0.375
!-0.316
인해-0.310
Furthermore-0.298
🤩-0.296