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 morning light, unusually bright, streamed through the kitchen window, illuminating dust motes dancing in the air. Amelia, a successful architect with a reputation for practicality, stared at the drawing tacked to her refrigerator. Her niece, Lily, had sketched a whimsical creature with mismatched wings and a mischievous grin. It was a familiar face. A face she hadn't seen in decades. A smile tugged at Amelia's lips, a rare and genuine expression.

She reached for her sketchbook, forgotten for months. The smooth, cool paper beckoned. Her fingers, usually steady with a drafting pencil, now trembled slightly. An idea, a vibrant bloom, began to unfurl within her.

She flipped through the blank pages, the world suddenly bursting with possibilities. It felt like her mind was buzzing, a hive of potential. She grabbed her pencils and started, her strokes rapid, almost frantic, as she attempted to capture the feeling, the echo of a forgotten playmate.

Emotion: inspired

Cluster: Positive / Joy
PC1 (Valence): 2.98 Positive
PC2 (Disposition): 0.04

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.563
invigor0.465
newfound0.446
هذ0.414
используя0.406
Suppressed:
😞-0.481
不-0.431
worse-0.428
😣-0.400
fact-0.389