© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129816
    Prev
    Next
    INDEX
    Explanations

    The neuron is detecting leading whitespace indent tokens (i.e. code indentation).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    programming code and names

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     подготовка
    -0.92
     Meksiko
    -0.90
     mungkin
    -0.89
     komment
    -0.86
    🕐
    -0.85
    PointF
    -0.82
     suaminya
    -0.82
     beker
    -0.81
    IENTS
    -0.81
    Ār
    -0.81
    POSITIVE LOGITS
     subida
    0.84
    sizable
    0.84
    oteo
    0.79
     технологі
    0.79
     use
    0.79
    (""+
    0.79
    corrosion
    0.77
     росій
    0.77
    MMV
    0.77
    Using
    0.76
    Activations Density 0.002%

    No Known Activations