© 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. 129430
    Prev
    Next
    INDEX
    Explanations

    The neuron is picking up on long runs of leading whitespace (indentation) at the start of code lines.

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

    C++ and Java code syntax

    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
    ences
    -0.90
    SaveChangesAsync
    -0.84
    feel
    -0.83
     塗装
    -0.83
    teenth
    -0.82
    van
    -0.82
    ㎡
    -0.81
    قي
    -0.81
    Wondering
    -0.81
    guys
    -0.81
    POSITIVE LOGITS
     richtigen
    0.86
     acquisitions
    0.86
    機場
    0.82
     miglior
    0.82
    涂
    0.81
     cotone
    0.80
    ayah
    0.80
     Bewegungen
    0.80
     Mischung
    0.80
     pronounce
    0.79
    Activations Density 0.004%

    No Known Activations