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 surveyor's flag, a bright orange, flapped mockingly in the wind, marking the boundary line. Liam stood on his porch, fists clenched, as his neighbor, Mr. Davies, explained his reasoning. "It's a hazard," Mr. Davies said. "Damaging my property."

Liam pictured the tree's sturdy trunk, the place where he and his wife, Clara, had carved their initials decades ago. Their names were hidden now, but they were still there.

The image brought a lump to his throat. He shook his head slowly. He had lost Clara two years ago, and the tree felt like a last link to her, to the promise of forever they had shared. “I don’t want it cut down," Liam rasped.

Emotion: sentimental

Cluster: Nostalgia / Reflection
PC1 (Valence): 0.89 Positive
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.677
fondly0.567
memories0.503
曾0.488
reminis0.466
Suppressed:
either-0.515
or-0.488
[]-0.473
--0.408
не-0.396