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 sales meeting dragged on. The air conditioning was struggling, and the speaker’s voice was like nails on a chalkboard. Then, to her dismay, it was her turn.

“Anyone remember Camp Greenleaf?” David asked casually, leaning back in his chair.

Her heart sank. She hated David. She hated Camp Greenleaf.

“Oh, yes,” she replied, her voice tight. “Such a classic.” She looked around at everyone, pretending to care.

“I was the archery champion that year.” David smiled, beaming.

She took a deep breath. “That’s… wonderful.” She then spoke, forcing a grin. “However, let’s move on to the sales figures for Q3…” She launched into the numbers, her voice clipped, barely looking at him.

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