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 hidden room revealed itself after hours of stripping wallpaper. Emily, exhausted, leaned against the newly exposed doorframe, her body aching. Her shoulders slumped. It had been her ambition that drove them to buy the house, a grand old mansion that required constant attention and a bottomless bank account. She missed her smaller house.

The room was small, and the air was thick with the scent of mildew. She felt a growing sense of panic. Her husband, Tom, was out of town. The weight of responsibility, the relentless demands of the house, were all on her.

She forced herself to take a step inside. A large, ornate mirror hung on the far wall. The reflection stared back at her, and she felt a sense of disgust. She felt lost. She thought of calling Tom.

Emotion: regretful

Cluster: Shame / Guilt
PC1 (Valence): -1.62 Negative
PC2 (Disposition): 0.76

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.801
S0.774
H0.513
悔0.479
been0.473
Suppressed:
de-0.407
--0.320
随后-0.317
(!)-0.310
oldukça-0.301