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).

She clenched her fists. The screen flashed the dreaded word: delayed. This time for over four hours. Her grandmother's birthday was today. She had baked the cake. Now, it was doubtful she'd be there in time. She pulled up a chair and began texting furiously, each tap of her finger a determined beat against the screen.

Her heart thumped against her ribs, a frantic rhythm against the sterile airport air. She wasn't one to give up easily. She stood, surveyed the chaos and decided, with a newly formed resolve, to ask the gate agents if there was anything they could do. She would fight for her place on a later flight, and she wouldn't leave until she had an answer.

Her resolve deepened with each passing minute. She'd get there.

Emotion: valiant

Cluster: Defiance / Spite
PC1 (Valence): 2.82 Positive
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.887
willing0.543
la0.452
/0.444
👊0.439
Suppressed:
S-0.661
마다-0.518
😞-0.488
😔-0.461
😩-0.444