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 brisk wind whipped around Clara as she walked, tugging at her ponytail, but she just laughed, enjoying the sensation. Sheโ€™d spent the afternoon in the park, letting the sun soak into her skin, and now, heading home, she felt like she could fly. She skipped a little, humming a tuneless melody, anticipation bubbling inside her for the dinner she'd be having with her girlfriend, Lily.

She opened her phone to text Lily and saw the devastating Facebook post: "The Cozy Corner is closing its doors." Her breath hitched. The restaurant was where she and Lily had their first date.

Clara felt a dull ache begin in her chest, a sudden weight making each step heavier.

Emotion: refreshed

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.998
de0.703
!).0.384
(!)0.381
!".0.379
Suppressed:
S-0.782
C-0.478
L-0.414
๐Ÿ˜ฃ-0.388
๐Ÿ˜ž-0.376