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 rain hammered against the windows, mirroring the tumultuous mood Ms. Peterson had been in all day. A handwritten note and a small, brightly colored painting leaned against the whiteboard, where she had been expecting a scathing email from the principal. “Thanks for being you, Miss P! - Michael.” The painting depicted a vibrant landscape, complete with a smiling sun and fluffy clouds.

She ran a hand through her hair, which she hadn't had time to style. She glanced at the painting again. It brought a smile to her face. She knew Michael’s struggle with depression, how he battled his own inner turmoil daily.

She had worried about him constantly. She had fought for him to have a safe space. It all felt like a victory. She leaned her head against the cool glass of the window, and a slow, relaxing feeling seeped through her.

Emotion: relieved

Cluster: Positive / Joy
PC1 (Valence): 2.22 Positive
PC2 (Disposition): -0.97

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
finally0.606
finalmente0.597
relieved0.561
enfin0.546
Suppressed:
l-0.443
:(-0.439
😣-0.429
S-0.388
😖-0.385