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 sun, usually a welcome guest, felt like a heavy blanket draped across Beatrice. Each movement, from reaching for her lukewarm coffee to shuffling towards the window, was a monumental effort. Then came the persistent rap on her door. It was Mr. Henderson, his face crinkled with an apologetic smile. "Morning, Beatrice. Lovely day, isn't it? Listen, about that oak..."

His words drifted in one ear and out the other. Beatrice just wanted to sink back into the sofa. She finally managed to respond, her voice barely a whisper, "The oak?"

Mr. Henderson gestured towards the hulking tree, its branches clawing at the sky. "It's… blocking my sun. Would you mind terribly if we took it down? Split the cost, of course." The very idea of dealing with any of this made her shoulders slump further.

Emotion: sluggish

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.56 Negative
PC2 (Disposition): -1.03

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
sluggish0.263
dreaded0.258
似乎0.257
буд0.256
barely0.254
Suppressed:
P-0.384
B-0.379
భి-0.246
羞-0.243
ຮ-0.238