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 movers were gone, and the echo of their footsteps had finally faded. David surveyed his new apartment. It was the perfect blend of modern and cozy, and he'd finally be able to focus on his writing without the distractions of the city. He stretched out his arms, a sigh escaping his lips. Everything just felt right.

He was checking the back of his bedroom closet when he felt a draft of air. There was a faint click. Curious, he pressed against the back wall. It gave way, revealing a narrow staircase leading downwards.

He felt a grin creep across his face as he took the stairs. The room was simple, bare, and lit by a single bulb. But it was *his*. Another secret space to make his own.

Emotion: happy

Cluster: Positive / Joy
PC1 (Valence): 3.98 Positive
PC2 (Disposition): -1.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.676
excitedly0.533
enthusi0.505
de0.498
๐Ÿคฉ0.495
Suppressed:
S-0.896
๐Ÿ˜ž-0.430
ness-0.410
ใใ‚Œใงใ‚‚-0.387
L-0.379