© 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. 77085
    Prev
    Next
    INDEX
    Explanations

    Java class loading or templates

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

    This neuron detects code-style tokens and XML/HTML tags—i.e. strings containing uppercase letters, punctuation, or non-alphabetic symbols (class names, macros, tags, file extensions, etc.).

    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
    AMPLE
    -0.77
    珍
    -0.77
     intermission
    -0.75
     ಕ
    -0.71
    engine
    -0.69
    mogat
    -0.68
    secure
    -0.68
    xmlhttp
    -0.67
     GREEN
    -0.66
    license
    -0.65
    POSITIVE LOGITS
    acyl
    0.73
    ھر
    0.69
    Ff
    0.67
    dion
    0.66
    Hh
    0.65
    bé
    0.64
     被
    0.63
     perempuan
    0.62
    DateTimeFormat
    0.61
    úmero
    0.60
    Activations Density 0.060%

    No Known Activations