© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21434
    Prev
    Next
    INDEX
    Explanations

    introduces programme, board, URL

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

    The neuron fires on the single key “action” word (usually the main verb) in titles or lead‐in lines (e.g. introduce, set, is, are).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     içerisinde
    -1.63
     fuimos
    -1.57
     tuve
    -1.57
     hermosas
    -1.55
     thậm
    -1.48
     тебя
    -1.48
     maybe
    -1.45
     utilizamos
    -1.45
    ívne
    -1.41
     your
    -1.41
    POSITIVE LOGITS
    eration
    1.67
    💐
    1.47
    beams
    1.42
    💓
    1.42
    +;
    1.38
    💞
    1.36
    桖
    1.34
     to
    1.33
     flot
    1.31
    ,
    1.30
    Activations Density 0.344%

    No Known Activations