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 archaeologist, Maya, brushed the dust from the ancient artifact, her hands trembling with a mixture of fatigue and anticipation. The inscriptions, she believed, held the key to unlocking a lost civilization. She was so close.

A young man, David, approached her, his face alight with excitement. He held a sketchpad, filled with remarkably similar drawings and translations. “I think we’re on the same page,” he said, his voice hushed with reverence. "The same symbols…the same secrets."

Maya felt a lightness in her limbs, a lightness in the air. The weight of solo work shifted to something else. Her heart seemed to expand. The past, and the future, felt less enigmatic.

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