© 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. 23614
    Prev
    Next
    INDEX
    Explanations

    edit mode

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

    The neuron is looking for code‐style keywords and identifiers (e.g. programming/API terms like “edit,” “dialog,” “item,” “DetailsViewMode,” etc.).

    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
    Signing
    -0.84
    也許
    -0.78
    baan
    -0.77
    inghouse
    -0.73
    LoginActivity
    -0.71
    irce
    -0.71
     sire
    -0.71
     ဝ
    -0.70
    Instal
    -0.70
    bane
    -0.70
    POSITIVE LOGITS
     edit
    2.11
     editing
    2.03
     Edit
    1.77
     EDIT
    1.75
    edit
    1.74
    EditMode
    1.72
    Edit
    1.61
    EDIT
    1.52
     edits
    1.52
    editing
    1.50
    Activations Density 0.017%

    No Known Activations