© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 73970
    Prev
    Next
    INDEX
    Explanations

    LEAD Summit, SMART Program, Reset the Net

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

    The neuron fires on multiword campaign or program titles—especially brief, upbeat action‐style names (e.g. “Stand Up,” “Power Up,” “Invest In Africa,” 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_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
    Efter
    -1.73
     desglose
    -1.55
     câbles
    -1.49
    Berikut
    -1.48
     fijar
    -1.41
     erwähnten
    -1.38
     réfrig
    -1.35
     好看
    -1.34
     gavetas
    -1.34
    Опис
    -1.33
    POSITIVE LOGITS
     them
    1.77
     in
    1.71
     for
    1.59
     not
    1.52
     into
    1.48
    us
    1.34
     "
    1.31
    /
    1.29
    ne
    1.28
     mereka
    1.26
    Activations Density 0.129%

    No Known Activations