© 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. 57192
    Prev
    Next
    INDEX
    Explanations

    parser, scanner, tokens

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

    The neuron fires on code‐style identifiers and symbols—i.e. variable, function, module, and other programming names (including underscores) in source code.

    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
     зависи
    -0.75
     voud
    -0.75
    orithmic
    -0.74
    inalg
    -0.72
    ongo
    -0.72
     tf
    -0.69
     rural
    -0.69
    ichts
    -0.68
     revistas
    -0.67
     modelo
    -0.66
    POSITIVE LOGITS
     yy
    1.77
    YY
    1.57
     YY
    1.49
    yy
    1.45
    bison
    1.02
     lex
    1.01
    Yy
    0.99
     yaka
    0.96
    yac
    0.93
     y
    0.88
    Activations Density 0.066%

    No Known Activations