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 attic air tasted of dust and forgotten things. Elias, knees aching, reached for the cardboard box labeled "Mom's Things." Inside, nestled amongst moth-eaten shawls and yellowed photographs, was a worn leather-bound notebook. He flipped it open, the brittle pages whispering in the quiet room. A handwritten list stared back at him: "See the Northern Lights. Learn to bake a perfect sourdough. Visit the Grand Canyon..." A smile tugged at the corners of his mouth. This was going to be interesting. He stood and stretched, a sudden lightness in his limbs. "Well, Mom," he murmured, his voice thick with unbidden emotion. "Let's get started."

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