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 aroma of freshly baked bread filled the air as Martha saw Mrs. Chen approaching, a plate in her hand. A peace offering, or a prelude? Her stomach clenched.

“Martha, dear, I brought you some cookies! And, about that cherry tree…” Mrs. Chen’s smile was wide, but her eyes held a calculating glint.

The cherry tree. The source of so much beauty, so much… longing. Martha felt a strange, suffocating tightness in her chest.

“The fruit is making such a mess of my patio,” Mrs. Chen said, her voice laced with a subtle pressure. Martha's hands, usually steady, began to tremble.

She took the plate, forcing a smile. She was trapped, a moth pinned under glass, at the mercy of the elements.

Emotion: vulnerable

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.327
ness0.319
😞0.299
😔0.277
😣0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257