© 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. 99819
    Prev
    Next
    INDEX
    Explanations

    script license or copyright

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

    The neuron fires on code‐style property declarations—tokens forming property names with punctuation separators (e.g. CSS hyphenated names like “list-style:” or object member accessors like “Manager.ConnectionStrings”).

    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
    wixt
    -0.79
    tools
    -0.79
    itesse
    -0.71
     portanto
    -0.70
    denas
    -0.69
     units
    -0.69
     Zone
    -0.67
     Dink
    -0.67
    tici
    -0.66
    gz
    -0.66
    POSITIVE LOGITS
    banned
    0.71
     Pah
    0.69
    Pah
    0.68
     decl
    0.65
    וסט
    0.65
    decrypted
    0.64
    Recipient
    0.63
     cave
    0.63
    univer
    0.62
    longValue
    0.62
    Activations Density 0.072%

    No Known Activations