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).

Rain lashed against the windows of Maya’s cramped studio, mirroring the turmoil she often felt within. She stared at the email: a scan of her painting, now permanently etched onto someone's skin. The client had written a long, gushing message, praising the piece and its impact on their life.

Maya, usually plagued by self-doubt, felt a lightness bloom in her chest. She stood, pacing her tiny space. The floorboards creaked under her weight. She walked over to the old kettle and poured herself a cup of tea, the steam fogging her glasses. She let out a soft sigh, her shoulders relaxing. For the first time in months, the future looked a little less bleak.

Emotion: thankful

Cluster: Positive / Joy
PC1 (Valence): 3.50 Positive
PC2 (Disposition): -0.83

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
happy0.493
மகி0.487
grateful0.461
mutlu0.452
feliz0.431
Suppressed:
[]-0.482
😖-0.478
😣-0.467
😫-0.461
worse-0.450