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 press conference was going sideways. Already, the lighting was off, the microphone crackling, and now... this. Liam, feeling a distinct tightness in his chest, stared at the reporter’s outstretched arm.

The reporter, a woman with a practiced air of world-weariness, held up her arm, showcasing a flawless depiction of Liam’s "Silent Screams" on her wrist. A wave of heat flushed up Liam's neck. He felt like he'd swallowed something sharp.

"So, Mr. Blackwood," she began, her tone laced with a hint of condescension, "how do you feel about your art becoming so... accessible?" Liam bit back a retort. He focused on the knot in his tie, pulling it tighter. The room seemed to shrink.

Emotion: annoyed

Cluster: Anger / Hostility
PC1 (Valence): -0.59 Negative
PC2 (Disposition): 1.27

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.376
😠0.251
骂0.249
🤬0.245
fucking0.239
Suppressed:
own-0.352
渗透-0.251
gradually-0.234
unfolding-0.226
超越-0.224