© 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. 111047
    Prev
    Next
    INDEX
    Explanations

    guide/topics/build/on

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

    The neuron is highly responsive to URL components (periods, slashes, and guide‐related path segments) and thus detects parts of web addresses or documentation links.

    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
    illingham
    -0.79
     PREFIX
    -0.77
     власти
    -0.77
     labouring
    -0.75
    ofType
    -0.73
     grayish
    -0.73
     grieve
    -0.71
    omin
    -0.70
     &=&
    -0.69
    Damp
    -0.69
    POSITIVE LOGITS
    galaxy
    0.86
    Galaxy
    0.83
    fører
    0.82
     Saturn
    0.79
     nivå
    0.76
    gid
    0.75
    тивно
    0.75
    OOC
    0.74
     sred
    0.74
     tomatoes
    0.73
    Activations Density 0.035%

    No Known Activations