© 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. 22-GEMMASCOPE-RES-131K
    7. 36340
    Prev
    Next
    INDEX
    Explanations

    keyword

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

    The neuron detects occurrences of the literal token “keyword” (in any capitalization or code‐context).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    {
    -3.13
    ↵
    -2.86
    ."
    -2.78
     The
    -2.42
    ’.
    -2.39
     It
    -2.36
     "
    -2.36
    .’
    -2.28
    其余
    -2.25
    {
    -2.23
    POSITIVE LOGITS
     estadounid
    2.63
    ྅
    2.55
     pérd
    2.42
    婓
    2.42
     复古
    2.36
     広場
    2.36
    螽
    2.34
     strategis
    2.34
     エントランス
    2.28
     fald
    2.27
    Activations Density 0.004%

    No Known Activations