© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21385
    Prev
    Next
    INDEX
    Explanations

    extramarital affairs and mistresses

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

    The main thing this neuron does is find words and phrases relating to infidelity or extramarital affairs.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     fantástico
    -1.63
    ？】
    -1.55
    取り組
    -1.48
     mágico
    -1.45
     duradero
    -1.41
     delicado
    -1.41
    ！】
    -1.39
    ～～～
    -1.34
     encantador
    -1.34
    者を
    -1.34
    POSITIVE LOGITS
     而
    1.46
    c
    1.36
    s
    1.29
    OTHER
    1.28
     mistress
    1.23
    3
    1.23
    g
    1.22
     ...
    1.20
     beauti
    1.20
     But
    1.18
    Activations Density 0.042%

    No Known Activations