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).

Sunlight streamed through the dusty attic window, illuminating motes dancing in the air. Eleanor, perched precariously on a rickety stool, dusted a forgotten trunk. Inside, nestled among moth-eaten fabrics, was a small, metal box. She tilted her head, a soft hum escaping her lips. "Well, isn't this interesting?" she murmured, her fingers tracing the tarnished surface. A wide grin stretched across her face.

Carefully, she lifted the lid. The contents were surprisingly intact. A photograph of a younger, vibrant her, holding a kitten. A childish, crayon-drawn picture of a unicorn. And a letter, addressed to her own, future self. Her breath hitched.

Her pulse quickened as she unfolded the letter. The familiar, looping handwriting was unmistakably hers. She leaned back, a small chuckle rumbling in her chest. This was a treasure, a little secret from a forgotten moment, just for her.

Emotion: pleased

Cluster: Positive / Joy
PC1 (Valence): 3.62 Positive
PC2 (Disposition): -0.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.728
la0.496
happy0.462
🤩0.433
🥳0.430
Suppressed:
S-0.486
😞-0.481
ness-0.454
😣-0.431
worse-0.401