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 banquet hall buzzed with forced cheer. Sarah fidgeted with the microphone, the cold metal a stark contrast to the clammy palms she couldn't seem to dry. Her mother, radiating joy in a shimmering emerald dress, beamed from the head table. Sarah swallowed hard, feeling a peculiar hollowness in her chest. She wished for a friendly face, someone to share this with besides the sea of unfamiliar relatives. She hadn't seen her sister in months, and felt a pang of longing for her presence.

The speech she had prepared felt inadequate, a series of rehearsed anecdotes that failed to capture the complexity of their shared history. It was all a bit… empty. She cleared her throat, the sound amplified and echoing in the room.

The cake sat like a mountain in the center of the room. It was huge. The thought of it made her stomach ache.

Emotion: lonely

Cluster: Sadness / Despair
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -1.29

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.416
loneliness0.375
孤独0.362
solitude0.351
lonely0.344
Suppressed:
de-0.450
l-0.360
/-0.272
👏-0.262
B-0.241