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).

He moved cautiously, afraid to breathe too loudly. His mother had passed away a year ago, but he was finally ready to sift through her things. He expected to find letters, photographs, keepsakes. He did not expect this.

Hidden in the back of a closet, nestled between blankets, were the relics of his childhood. Every single report card, every book report, every hand-drawn card – all neatly tucked away. A wave of intense emotion overwhelmed him as he held a faded drawing of himself.

A feeling of ease settled upon him. He knew he wasn’t supposed to be here, but his presence felt appropriate. He would be okay.

Emotion: hope

Cluster: Positive / Joy
PC1 (Valence): 3.59 Positive
PC2 (Disposition): -0.12

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.065
la0.865
B0.465
🤩0.412
H0.407
Suppressed:
😞-0.459
C-0.451
不-0.449
😣-0.443
S-0.442