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 anticipation of seeing Jackson bubbled within her. Evelyn was thrilled, giddy even. She had carefully chosen her outfit, a flowing dress that reflected the warmth she felt inside. She was meeting him at their favourite park, eager to share the story of her latest art commission. She had been a little down, but Jackson always knew how to make her smile. As she walked towards the park, she saw Jackson. He stood near a luxury car, holding a briefcase. The name on the car door read: "President, GreenLeaf Investments." Evelyn’s breath hitched. Jackson, who had always said he was a struggling artist himself. The day, which had promised sunshine, now felt overcast.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306