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 afternoon light glinted off the old, worn teddy bear sitting on the table. It was Barnaby, her oldest friend. Emily felt a sudden, inexplicable shift inside her, a feeling akin to coming home.

She reached out, gently touching his soft ear. Her fingers traced the worn patches, and she felt a wave of memories wash over her. It was a tangible connection to a happier, simpler time. A small smile crept onto her lips.

“How much?” she asked, her voice soft. She paid the price, then cradled Barnaby close as she left, a profound sense of peace settling over her. She knew she'd never let him go again.

Emotion: grateful

Cluster: Positive / Joy
PC1 (Valence): 3.77 Positive
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.571
de0.562
B0.546
மகி0.517
happy0.493
Suppressed:
[]-0.582
😖-0.519
worse-0.509
😫-0.498
😣-0.494