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 sun beat down on the empty apartment, its harsh rays highlighting every speck of dust. Daniel, sweating profusely, sifted through his grandfather's belongings. His grandfather had been a pillar of strength, a constant in Daniel's life. He had organized the old man's things when he passed, though the task always made him ill at ease.

He found a small, ornate box hidden beneath the wardrobe. Inside, nestled amongst cuff links and a worn pocket watch, was a letter. It was a draft resignation letter from his grandfather’s military service, dated shortly after his wedding. The letter spoke of a crushing pressure, of feeling a lack of agency, of a deep-seated unease that had permeated his life.

Daniel’s breath hitched in his chest. He always imagined his grandfather in uniform. The man was a symbol of strength and self-reliance. He read the final, rewritten line: “I am no longer sure I can.” He felt a sudden, suffocating feeling. His grandfather had always made decisions. He had always guided Daniel's choices. He reached for the phone. He needed someone, anyone, to tell him what to do.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258