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 rusted metal box felt strangely familiar under Elara's gloved hands. She brushed the dirt away, revealing the inscription: "To Future Me, From Me." A silly phrase, she thought, chuckling. The afternoon sun warmed her face as she pried the lid open, a little click echoing in the quiet park. Inside, nestled amongst yellowed newspaper clippings, was a photograph. A younger Elara, beaming, held a small, fluffy dog. A smile blossomed on her own face, a sensation of lightness filling her chest. The dog was long gone, but the feeling, the bright optimism in that old photo, was still there, a tangible connection to someone she’d forgotten. She felt herself wanting to sit and study the picture.

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