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).

"Well, this is underwhelming," Maya muttered, staring at the chipped teacup. Her brother, Daniel, was the proud owner of a fully renovated Victorian mansion, complete with a swimming pool. She had received a single, fragile piece of porcelain, a remnant from their grandmother's forgotten tea service.

She picked it up, feeling the unevenness of the glaze under her fingertips. She longed for a sense of purpose. Ever since graduating college, she had floated from one temporary job to the next, unsure of where she belonged. The teacup felt like a physical representation of her aimlessness, a broken thing she didn't know how to mend.

She glanced at the perfectly manicured lawn of her brother's new home from the window. The contrast was a painful jab. She clutched the cup tightly.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263