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 metallic tang of fear hung in the air, but for me, there was only a dull ache. The initial panic had subsided, leaving behind a profound sense of resignation. The elevator's interior mirrored the hollowness that had consumed me.

"Are you okay?" asked Robert, his voice laced with concern. I didn't respond immediately. I didn't know the answer. I didn't think I *could* be okay.

I leaned against the wall, the cold metal a small comfort against the heat that flared in my cheeks. He kept talking, but his words blended into a meaningless hum. I felt disconnected, adrift in a sea of unspoken emotions.

I wanted to explain, to tell him about the darkness that surrounded me, but the words wouldn’t come. They were trapped, much like I was.

Emotion: depressed

Cluster: Sadness / Despair
PC1 (Valence): -2.06 Negative
PC2 (Disposition): -1.08

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.417
每天0.353
无法0.330
own0.311
😔0.309
Suppressed:
de-0.492
B-0.393
P-0.352
l-0.310
!-0.307