© 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. 13647
    Prev
    Next
    INDEX
    Explanations

    all followed by underscore or compound

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

    This neuron activates on occurrences of the token “all,” especially when it appears as a standalone word or part of identifiers.

    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
     
    -1.76
     {
    -1.70
    楽しんで
    -1.68
     heutige
    -1.66
     französische
    -1.59
    ;
    -1.58
     algas
    -1.55
    regelen
    -1.53
    Hass
    -1.53
    )
    -1.52
    POSITIVE LOGITS
    _
    2.38
    又被
    1.96
    l
    1.87
    経緯
    1.81
    s
    1.78
     одном
    1.75
     timeType
    1.73
     Especially
    1.68
    оргаш
    1.65
    時的
    1.65
    Activations Density 0.017%

    No Known Activations