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).

Rain lashed against the windows, mirroring the tempest inside Elara. She scrolled through her phone, the digital connection she’d fostered for months with "Nova" suddenly feeling like a cruel joke. Nova, who understood her sarcastic humor, who knew the names of her cats, who made her laugh even when the city felt like a concrete prison. And now? A quick search revealed her online confidante lived… five blocks away. Elara slammed her phone on the counter, the action fueled by a simmering resentment.

She paced her small apartment, a familiar ache spreading through her chest. The thought of running into Nova at the grocery store, at the park, anywhere, made her stomach churn. The shared secrets, the late-night chats… all tainted now. She’d crafted this persona, this safe space, and Nova, of all people, had infiltrated it, only to be *so close*.

The email notifications popped up, mocking her with the familiarity of Nova's name. They were filled with playful emojis and inside jokes, reminders of a connection that felt like a betrayal now. Elara slammed her laptop shut. The digital world she had created felt like a flimsy shield against the ever-present reality of the city she couldn't escape.

Emotion: bitter

Cluster: Shame / Guilt
PC1 (Valence): -1.78 Negative
PC2 (Disposition): 0.84

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.447
striving0.238
😞0.236
aggravated0.225
soport0.223
Suppressed:
own-0.316
de-0.291
la-0.231
🤩-0.227
是非常-0.225