© 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. 34-GEMMASCOPE-RES-131K
    7. 83215
    Prev
    Next
    INDEX
    Explanations

    administration paths

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

    navigation paths within control panels, admin interfaces, and file system directories.

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

    This neuron detects navigational or location‐directive language in technical instructions—phrases and tokens like “go to,” “in,” “under,” menu or file‐path segments (e.g. “admin,” “/structure/…”).

    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
     infectious
    -0.84
    LoginComponent
    -0.84
    kategori
    -0.82
    DialogComponent
    -0.81
     wre
    -0.79
     Exact
    -0.79
    気がする
    -0.78
    Primera
    -0.77
    に関する
    -0.76
    そういった
    -0.76
    POSITIVE LOGITS
    Administration
    0.97
    Admin
    0.93
     Administration
    0.90
    भा
    0.88
     your
    0.86
     администра
    0.83
    首页
    0.83
    administration
    0.83
     left
    0.81
    tése
    0.81
    Activations Density 0.049%

    No Known Activations