© 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. 34-GEMMASCOPE-RES-131K
    7. 64618
    Prev
    Next
    INDEX
    Explanations

    code comments and copyright notices

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

    The neuron fires strongly on pieces of literal text (e.g. exception names, code strings, include/import paths)—i.e. it picks out tokens inside string‐literal or include‐directive quotes/brackets.

    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
    及其
    -0.77
     not
    -0.76
     camaraderie
    -0.75
    hamdulillah
    -0.73
    erdings
    -0.72
    ="/"
    -0.70
    sä
    -0.70
    스템
    -0.70
     wanna
    -0.69
     samurai
    -0.69
    POSITIVE LOGITS
    Against
    0.99
     ugyan
    0.93
     against
    0.92
     añadió
    0.92
     nämlich
    0.90
    */
    0.90
     namelijk
    0.90
    heria
    0.90
     추가
    0.89
    return
    0.88
    Activations Density 0.004%

    No Known Activations