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 sunshine felt wrong today. Too bright, too cheerful. The unsigned card, slipped anonymously into Eleanor's mailbox, felt like a deliberate intrusion. A deliberate jab. She could almost taste the smugness of the sender.

She examined it, the saccharine sentiment doing nothing to allay her mood. Eleanor slammed it onto her kitchen counter. A quick, sharp thought occurred to her, and she smiled. She rummaged through her drawers until she found what she needed.

She spent the afternoon concocting a batch of her infamous chili. Extra hot. Extra spicy. Then, she casually placed a small, unmarked container on the doorstep of her primary suspect. Let them enjoy their day.

Emotion: spiteful

Cluster: Defiance / Spite
PC1 (Valence): 2.20 Positive
PC2 (Disposition): 3.95

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.717
l0.517
a0.480
la0.440
😈0.428
Suppressed:
own-1.063
此刻-0.534
గుర్త-0.493
熟悉的-0.487
ness-0.479