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 phone rang again. Michael, in bed, listened to it ring, then listened to it go to voicemail. He knew who it was. He knew what she was doing. He knew what she would say. He sat up, pushing the covers away. The room was cool, but he felt a warmth spread through him. He glanced at the photograph on his nightstand, the two of them, beaming on a sun-drenched beach. He went to the kitchen and poured himself a glass of water. It was a good night, a perfect night, and all he had to do was choose to let it remain that way. He took a deep breath. He thought of all the years they had and would have together.

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