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 glow of the computer screen illuminated Thomas's face as he scrolled through the online bank statements. His wife, Emily, had a secret account, a nest egg built slowly, steadily, and without his knowledge. He clicked the delete button on the email, but the betrayal was not erased. His jaw was set.

He poured himself a stiff drink, pacing the living room. When Emily came home, humming, her arms filled with grocery bags, he offered her a tight smile. "You know, the washing machine is on its last leg," he began, his voice surprisingly calm. "We really need a new one." Emily sighed, explaining about the current month's budget.

"But what about that trip to Italy we were talking about?" he probed, a deliberate edge creeping into his tone. "I thought we could finally book it." Emily’s smile wavered. He watched her, a thin line across his lips, silently enjoying her discomfort.

Emotion: stubborn

Cluster: Defiance / Spite
PC1 (Valence): 0.80 Positive
PC2 (Disposition): 3.73

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.780
l0.685
la0.605
either0.580
a0.501
Suppressed:
难以-0.435
终于-0.419
顿时-0.405
竟然-0.396
wondering-0.391