© 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. 78620
    Prev
    Next
    INDEX
    Explanations

    code setup class inheritance

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

    The neuron fires on file‐header and comment‐block metadata tokens—things like comment asterisks (“*”), code snippet markers (“CODE:” and “setUp”), and numeric header fields (dates, version numbers, case or license identifiers).

    oai_token-act-pair · o4-miniTriggered 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
     गु
    -0.68
    fehlung
    -0.66
    deviantart
    -0.64
     hãng
    -0.64
     مت
    -0.63
     America
    -0.63
     һәм
    -0.63
    信息
    -0.61
     epilepsy
    -0.61
    nude
    -0.60
    POSITIVE LOGITS
    JAC
    0.72
    ULAR
    0.71
     concession
    0.71
    רץ
    0.69
     Ome
    0.68
     Buss
    0.67
     perdre
    0.66
     Peek
    0.65
     NEST
    0.64
     Lehr
    0.64
    Activations Density 0.063%

    No Known Activations