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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21905
    Prev
    Next
    INDEX
    Explanations

    protect the innocent/guilty

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

    The neuron detects statements that personal names or identifying details have been redacted or changed to protect someone’s privacy.

    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
     pirata
    -1.68
     parati
    -1.39
     itin
    -1.36
     makam
    -1.34
     bilang
    -1.32
    opies
    -1.32
     tortura
    -1.30
     meras
    -1.30
     tuer
    -1.28
     turno
    -1.27
    POSITIVE LOGITS
     be
    1.71
     to
    1.34
     if
    1.30
     also
    1.30
    9
    1.27
    霄
    1.20
     there
    1.20
    憐
    1.20
    匿
    1.20
     from
    1.18
    Activations Density 0.028%

    No Known Activations