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 cramped office seemed to shrink slightly as Eleanor, the new junior analyst, settled in beside Marcus. He felt a light, almost giddy feeling as he turned to her, ready to begin the training. He, after all, was the go-to person on this stuff.

He launched into a detailed explanation of the companyโ€™s financial models, his voice calm and clear, explaining the nuances that only he, after all this time, really understood. The complex formulas, the intricate calculations โ€“ they all fell under his expert gaze. Each sentence came out with such precision.

He spoke with an air ofโ€ฆ well, authority. It was a good feeling, a kind ofโ€ฆ warmth. The feeling that all the long days and nights spent poring over spreadsheets had been worth it.

Emotion: proud

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.514
la0.457
๐Ÿ˜Ž0.400
happy0.394
๐Ÿ‘0.393
Suppressed:
S-0.672
๐Ÿ˜ž-0.480
๐Ÿ˜ฃ-0.460
worse-0.455
๐Ÿ˜–-0.453