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 attic door groaned open, spewing dust motes into the afternoon sunbeam. Mark bounced on the balls of his feet, his hands already reaching for the pull-cord that dangled from the light fixture. He couldn’t wait to see what secrets lay hidden. He’d barely unpacked the kitchenware before he was drawn to this spot, a whisper in the back of his mind promising forgotten treasures.

A thick layer of grime coated everything. Ignoring it, Mark plunged forward. Boxes of all shapes and sizes were stacked haphazardly, a testament to the lives lived before him. He practically vibrated with anticipation.

He grabbed a box labeled "Miscellany." It was heavy, and as he wrestled it open, a rush of stale air escaped. Inside, nestled amongst yellowed newspaper, was a collection of antique buttons. He let out a breathless laugh. This was going to be a good day.

Emotion: eager

Cluster: Positive / Joy
PC1 (Valence): 2.98 Positive
PC2 (Disposition): 0.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.659
excitedly0.436
!0.389
!"0.386
ecstatic0.383
Suppressed:
S-0.816
ness-0.441
😞-0.387
L-0.379
P-0.376