© 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. 34-GEMMASCOPE-RES-131K
    7. 63971
    Prev
    Next
    INDEX
    Explanations

    copyright notices and schemas

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

    The neuron fires strongly on standalone numeric tokens, effectively detecting whenever a number appears.

    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
    ец
    -0.81
    kont
    -0.79
     menghar
    -0.77
     вверх
    -0.75
    wij
    -0.74
     sidan
    -0.69
    obod
    -0.68
     swan
    -0.68
    atat
    -0.67
    ≼
    -0.66
    POSITIVE LOGITS
    caval
    0.73
     auszu
    0.68
    engin
    0.64
    uga
    0.63
    UGAL
    0.62
     benzene
    0.62
    ชาว
    0.62
    dentes
    0.61
    hoea
    0.60
     Penelope
    0.60
    Activations Density 0.062%

    No Known Activations