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 aroma of freshly baked bread wafted from the kitchen, a symphony of warmth. Sarah hummed along to the radio, kneading dough with a gentle rhythm that matched the beat. Her new sourdough starter, Gertrude, was proving to be a true joy.

A sharp rap echoed from the front door. Sighing, she wiped her hands on her apron and went to answer it. It was Mr. Henderson, his face contorted in a familiar grimace. "That infernal music," he began, his voice tight. "It's just too loud."

Sarah tilted her head, a small smile playing on her lips. "Oh, Mr. Henderson, I do apologize," she said, her tone light and airy. "I was just celebrating Gertrude's success, I’ll turn it down right away." She envisioned the perfect crust forming, the tangy flavor bursting forth – and the coming day, of sharing loaves with the people she loved.

Emotion: fulfilled

Cluster: Positive / Joy
PC1 (Valence): 3.22 Positive
PC2 (Disposition): -1.74

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.819
மகி0.415
happy0.397
grateful0.389
de0.385
Suppressed:
S-0.599
worse-0.465
C-0.460
either-0.436
😣-0.426