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).

He’d called, finally. That was the first thing. The phone vibrating against the desk, the ringtone's cheerful jingle mocked me. I let it go to voicemail, pretending to ignore it, even though I knew the message was probably from Liam. The silence felt heavy.

The apartment was a mess, reflecting the chaos inside. Dishes piled in the sink, clothes strewn across the floor. I usually kept things tidy. Now, the disarray matched my mood, a messy echo of my own inner turmoil. I grabbed a can of soda, popped the top with a violent hiss, and threw it into the nearest bin.

He called again. I finally grabbed the phone, my voice low and brittle.

Emotion: irritated

Cluster: Anger / Hostility
PC1 (Valence): -0.86 Negative
PC2 (Disposition): 1.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.370
S0.282
P0.264
😠0.248
aggravated0.233
Suppressed:
own-0.300
joyful-0.211
joyous-0.207
खोले-0.206
newfound-0.205