© 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. 22-GEMMASCOPE-RES-131K
    7. 33355
    Prev
    Next
    INDEX
    Explanations

    remove invalid text

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

    The neuron flags language about automatically filtering or stripping out invalid input characters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    駅から
    -1.05
     vélos
    -1.02
    的气息
    -1.02
     romero
    -1.02
     bâtiments
    -0.98
     animaux
    -0.97
     geluk
    -0.97
    老師
    -0.96
    instead
    -0.95
    vajú
    -0.95
    POSITIVE LOGITS
    埴
    1.10
    Lycka
    1.03
    /******/
    0.96
    umsy
    0.92
     cursor
    0.91
     麦
    0.91
    Varför
    0.91
     BEFORE
    0.91
    Mål
    0.90
     insertion
    0.90
    Activations Density 0.023%

    No Known Activations