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 air hummed with a strange energy as Daniel opened the dusty box. He’d been cleaning out the attic, a task he’d been putting off for months. At the bottom, a bound manuscript, its cover blank save for a single word: "Echoes."

He flipped the pages, recognizing himself in the first paragraph. He was the main character. He was the subject of her story. His chest felt tight with excitement and curiosity.

He sat cross-legged on the floor, the world shrinking to the pages in his hands. He was pulled into the story, consumed by it, desperate to know what happened next. Every sentence was a revelation, every detail a secret shared.

He had to tell her. He had to know why. He had to know if she was proud.

Emotion: eager

Cluster: Positive / Joy
PC1 (Valence): 2.98 Positive
PC2 (Disposition): 0.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.659
excitedly0.436
!0.389
!"0.386
ecstatic0.383
Suppressed:
S-0.816
ness-0.441
😞-0.387
L-0.379
P-0.376