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 florist had been cheerful, almost too much so, and as Amelia took the arrangement, a mix of sunflowers and asters, she found herself shrinking back in unease. She shut the door quickly, the click echoing in the silence.

She carried the flowers into the living room, a space usually filled with calm. Her hand shook as she set them down on the coffee table. The vibrant colors seemed to pulsate, reflecting her own inner turmoil. She felt trapped.

She sat on the sofa, clutching the cushions, and stared at the flowers. Her mind raced with questions, each one amplifying the sense of dread that had suddenly consumed her. She felt a cold sweat break out on her forehead. The room felt as though it was closing in.

Emotion: on edge

Cluster: Fear / Anxiety
PC1 (Valence): -1.77 Negative
PC2 (Disposition): -0.35

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😖0.228
😣0.224
worse0.223
urges0.216
느껴0.214
Suppressed:
de-0.514
la-0.391
a-0.366
B-0.332
L-0.283