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 musty smell of the basement clung to Ava as she examined the photographs. She held a portrait of her father's father, Thomas, a man she barely remembered, a man whose life had always been shrouded in mystery. The photo showed him standing next to a beautiful woman, a woman who wasn't her grandmother. They were embracing, laughing, their faces illuminated by a warm, inviting light.

Ava felt a rush of emotions – confusion, betrayal, a strange, inexplicable sorrow. The stories she'd been told about Thomas now seemed like fragments of a much larger story. Her voice was lost in a whisper as she looked at the photo. The history she thought she knew was crumbling around her.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238