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 sunlight seemed too bright in the kitchen, highlighting every speck of dust and imperfection. Clara slumped onto a chair, the metal cold against her skin. Her head throbbed, a dull, persistent ache that echoed her mood. David was whistling, a jaunty melody that grated on her nerves.

He was in the middle of preparing his trademark carbonara, a dish he usually cooked with such a flourish, his usual joy.

"Ready for a movie night?" he asked, not turning around. She simply stared at the floor.

Suddenly, he started talking, a torrent of words tumbling out in rapid succession. She recognized the words, of course. Her native tongue, Gaelic, was now being spoken with a practiced ease. Her mouth felt dry, her heart rate increased. She felt like she was trapped.

Emotion: dispirited

Cluster: Sadness / Despair
PC1 (Valence): -1.88 Negative
PC2 (Disposition): -0.98

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.398
S0.348
’0.325
ness0.320
无法0.308
Suppressed:
de-0.297
l-0.256
👏-0.251
sonra-0.240
aud-0.240