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 adoption website was open on Leo’s laptop, the screen reflecting the flickering candlelight. Rain lashed against the window, a soundtrack to the silence. He’d carefully filtered the search: hypoallergenic, small, and cuddly. Each image of a fluffy, whiskered face brought a small tremor of… well, it brought on a feeling that everything would be okay.

"Look at this one, Sarah!" he exclaimed, pointing to a photo of a tiny terrier mix with floppy ears. Sarah sighed, running a hand through her hair.

"Leo, we talked about this. We both travel a lot for work. A pet just isn't feasible." But even her usual firmness seemed diminished. He'd shown her a compromise – a pet that wouldn’t need daily walks.

He took a deep breath. "What about a rabbit? They're relatively low maintenance. We can get a really nice hutch and keep it in the living room." His smile widened, and he felt a flutter of anticipation in his chest.

Emotion: hopeful

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.715
la0.549
B0.468
🤩0.452
enthusi0.433
Suppressed:
S-0.603
不-0.401
😞-0.399
worse-0.394
C-0.386