© 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. 34-GEMMASCOPE-RES-131K
    7. 91826
    Prev
    Next
    INDEX
    Explanations

    mixed categories

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

    The neuron predominantly activates on numeric tokens and numeric literals (e.g. version numbers, IDs, measurements) in the text.

    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
     STD
    -0.75
    ประก
    -0.66
    chts
    -0.65
     princí
    -0.64
     стол
    -0.64
    spells
    -0.63
     こう
    -0.62
     NN
    -0.61
     Argus
    -0.61
    siers
    -0.60
    POSITIVE LOGITS
    洀
    0.71
    שו
    0.69
    SDF
    0.69
     RADIATION
    0.68
    ęż
    0.67
    osin
    0.65
    ropshire
    0.65
     чудо
    0.64
     سانتی
    0.64
    貴
    0.64
    Activations Density 0.065%

    No Known Activations