© 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. 8365
    Prev
    Next
    INDEX
    Explanations

    cult, personality disorders, manipulation

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

    The neuron is detecting mentions of abusive behavior and domestic abuse (including terms like “abuse,” “abusive,” “abuser,” “psychological manipulation,” etc.).

    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
     for
    -1.30
     scuole
    -1.26
     aprobó
    -1.21
     "",
    -1.13
     knowing
    -1.13
     squadre
    -1.12
     terutama
    -1.09
     تجمع
    -1.09
    2
    -1.09
    ms
    -1.07
    POSITIVE LOGITS
     читать
    1.24
     cáncer
    1.16
     him
    1.13
     İç
    1.13
     psychological
    1.09
    CROSSTALK
    1.09
     how
    1.06
     clases
    1.05
    CASCADE
    1.05
    fassung
    1.03
    Activations Density 0.049%

    No Known Activations