© 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. 34-GEMMASCOPE-RES-131K
    7. 79080
    Prev
    Next
    INDEX
    Explanations

    code comments and keywords

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

    The neuron is detecting numeric literals (numbers) in the text.

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

    code comment lines that contain test or build/run directives within source files.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    line breaks or newlines in code, particularly after comment markers.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Boo
    -0.88
     dön
    -0.84
    hau
    -0.84
    eses
    -0.84
    tobe
    -0.81
    wegen
    -0.81
     regeringen
    -0.81
    wisata
    -0.80
    ниться
    -0.80
    LLLL
    -0.80
    POSITIVE LOGITS
     using
    0.79
     having
    0.78
     використо
    0.77
     efficiently
    0.77
     Having
    0.73
    還有
    0.73
    母親
    0.72
    世間
    0.71
     ngọt
    0.71
     through
    0.70
    Activations Density 0.006%

    No Known Activations