© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 118952
    Prev
    Next
    INDEX
    Explanations

    command-line arguments and help

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

    The neuron strongly responds to programming “action/command” vocabulary—tokens like action, command, emit, import, register, load, module, api, help, etc., i.e. terms used in code for defining or invoking commands, actions, or CLI parameters.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    DockStyle
    -0.87
     appri
    -0.79
     noexcept
    -0.77
    hoea
    -0.77
    raim
    -0.74
    tler
    -0.73
    同じ
    -0.73
    5
    -0.72
     دست
    -0.72
    リル
    -0.71
    POSITIVE LOGITS
     argparse
    1.32
     choices
    1.09
     choice
    1.04
    choices
    1.03
     Argument
    1.00
    nargs
    0.94
     existing
    0.91
     존
    0.90
     Existing
    0.89
     Choices
    0.88
    Activations Density 0.010%

    No Known Activations