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).

Rain lashed against the windows of the cafe, mirroring the tempest brewing inside Mark. He shoved his half-eaten croissant around his plate. Every mouthful of food felt like a chore, a burden. He’d been trying to explain the relentless bone pain to his therapist for weeks. He was sure she didn't grasp the severity, the constant ache that never truly disappeared.

A woman entered the cafe and sat at the table next to him. She clutched a steaming mug. The scent of coffee, usually appealing, only intensified his existing unease. She sighed.

"This blasted weather," she said, her voice laced with a similar tone to his own. "Does anything ever feel *right*?"

"Tell me about it," Mark muttered, picking at a crumb.

She looked at him then, her eyes narrowing slightly. "Are you… here for the same reason I am?" she asked hesitantly.

He stared back, bewildered. "Are *you* also getting the injections?"

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