INDEX

Explanations

harmful sexual content

np_acts-logits-general · gemini-2.5-flash-lite

The neuron responds strongly to nouns describing violent acts or crime-related violence.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

mentions of violence and violent crime.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

words related to violence, crime, and harmful actions.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ון

0.84

 Flaxman

0.84

 नमी

0.83

으면

0.82

sheet

0.82

ällä

0.78

 naphthalene

0.77

 Ultrasonic

0.77

oload

0.76

POSITIVE LOGITS

 murders

1.13

 violent

1.12

 violence

1.10

 asesinato

1.06

 Violence

1.05

 killings

1.04

 perpetrated

1.01

 murder

1.00

violence

0.99

 atrocities

0.96

Activations Density 0.008%

harmful sexual content

The neuron responds strongly to nouns describing violent acts or crime-related violence.

mentions of violence and violent crime.

words related to violence, crime, and harmful actions.

No Comments

No Known Activations