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).

Liam paced his living room, the worn rug offering no solace under his feet. Heโ€™d just received the email from his daughter's school: Mr. Finch, Grade 2. The same Mr. Finch who'd called him "slow" and "unmotivated" in the same grade. Liamโ€™s hands clenched into fists, the knuckles white. He ran a hand through his hair, tugging at the strands as if trying to physically extract the past. He felt a deep, unsettling unease, a knot forming in his gut. His daughter, Chloe, was so young, so full of joy. He had to do something.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
๐Ÿ˜ž0.343
S0.317
ness0.302
๐Ÿ˜”0.294
๐Ÿ˜ข0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236