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 moment she saw him, her smile faltered. Maya's face, meticulously made up, felt like a mask. She’d been enjoying herself, chatting with friends, the joyful atmosphere of the wedding enveloping her. Now, the happiness had vanished, replaced by a deep disquiet.

He was standing near the buffet table, looking as handsome as she remembered, a small smile playing on his lips as he talked to her best friend. Maya's meticulously planned outfit suddenly felt too loud, too desperate. She felt exposed, vulnerable.

She took a step back, instinctively seeking the shadows. She wanted to be invisible. She busied herself with rearranging the flowers on a nearby table, her movements stiff and unnatural. Her heart pounded a frantic rhythm against her ribs.

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