© 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. 15082
    Prev
    Next
    INDEX
    Explanations

    porous substrates

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

    C/C++ syntax elements for structure/class declarations ending with semicolons.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on semicolon tokens—i.e. the “;” that ends declarations or statements.

    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
    彼らは
    -1.10
    haviours
    -1.06
     tráfico
    -1.05
    ZX
    -1.04
     balón
    -1.03
    疫苗
    -1.02
    reté
    -0.98
     fome
    -0.98
    vgl
    -0.93
    CHORUS
    -0.93
    POSITIVE LOGITS
     if
    1.21
    if
    1.09
    ]
    0.91
    开
    0.89
    \\
    0.86
    We
    0.85
    CMake
    0.85
     bound
    0.82
    </h3>
    0.81
    end
    0.81
    Activations Density 0.001%

    No Known Activations