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 will was read, the words echoing in the hushed, opulent drawing room. "To Beatrice," the solicitor announced, "the family mansion and the vast art collection." Beatrice, draped in a silk dress, subtly beamed.

Across the room, Arthur’s shoulders were rigid. He clenched his fists, knuckles white, as the man continued, “And to Arthur, a small, abandoned workshop in the city and a collection of tools.” It was what he'd expected.

The solicitor finished his reading. “The workshop will need considerable repair.” Arthur felt a rush of heat in his cheeks. He wasn’t going to beg for anything. He would prove them all wrong. He would fix that workshop and make something of himself.

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