© 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. 22-GEMMASCOPE-RES-131K
    7. 64979
    Prev
    Next
    INDEX
    Explanations

    code construct punctuation

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

    The neuron consistently lights up on identifiers that name or invoke functions or methods (i.e. tokens for function/method names before “(”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ]!='
    -0.94
    ridescent
    -0.94
     advierte
    -0.93
     concluye
    -0.90
     constituye
    -0.90
     Beweis
    -0.89
     ayudará
    -0.87
     demuestra
    -0.86
     acompanhar
    -0.85
     ativar
    -0.85
    POSITIVE LOGITS
     febbraio
    1.02
    oplastic
    0.96
    ٧
    0.95
    🤨
    0.88
    をしました
    0.86
    ができます
    0.84
    乙
    0.84
    ิ่ม
    0.84
    tare
    0.84
    ٨
    0.83
    Activations Density 0.004%

    No Known Activations