© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67963
    Prev
    Next
    INDEX
    Explanations

    command and code execution

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

    The neuron fires on code tokens that name or invoke external commands or string‐formatting expressions (e.g. COMMAND, EXECUTABLE, `${…}`, stringWithFormat, axios calls, subprocess.call, return).

    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
     bisous
    -1.16
    aturan
    -1.15
    ategorias
    -1.13
    anggap
    -1.13
     marki
    -1.09
    liebe
    -1.08
    utama
    -1.08
     poemas
    -1.06
    gunakan
    -1.02
     oferty
    -1.01
    POSITIVE LOGITS
    le
    1.07
     **/
    1.06
    is
    1.01
     was
    1.00
    ot
    0.96
     """
    0.96
    iden
    0.93
     petra
    0.93
     dokonce
    0.93
     espres
    0.93
    Activations Density 0.001%

    No Known Activations