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).

He flinched as he saw Mrs. Gable step onto her porch. The setting sun cast long shadows across the neglected grass. It highlighted the unkempt state of the yard. He felt a sudden, inexplicable urge to hide.

The woman was a master gardener, a source of constant wonder and envy for him. He, on the other hand, had let the yard fall into disrepair, a testament to his own shortcomings.

She was walking towards the fence, a small trowel in her hand. He imagined the way his side looked to her, a wasteland compared to her cultivated paradise.

"Good evening, Mr. Johnson," she said, her voice polite. "I was thinking about planting some lavender along the fence. Would you mind if I perhaps…" He interrupted her, his voice hoarse. "Sure, whatever you need." He turned away abruptly, unable to meet her gaze.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242