© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 109332
    Prev
    Next
    INDEX
    Explanations

    address bar and search box

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

    The neuron detects instructional phrases that tell the user to input or paste text into interface elements (e.g. “into the address bar,” “in the membership ID field,” “into an HTML form,” 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
    र्स
    -0.81
     listado
    -0.77
     победы
    -0.75
    isContained
    -0.74
    ньше
    -0.71
    hasMoreElements
    -0.71
    ถือ
    -0.71
     chooses
    -0.69
    子供が
    -0.69
    removing
    -0.69
    POSITIVE LOGITS
     field
    2.80
     fields
    2.66
     box
    2.41
     boxes
    2.08
     space
    1.96
     поле
    1.89
    fields
    1.88
    Fields
    1.87
    field
    1.86
    Field
    1.84
    Activations Density 0.087%

    No Known Activations