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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 60256
    Prev
    Next
    INDEX
    Explanations

    Jessica Morgan, Role, Policy

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

    The neuron is selectively activating on numeric tokens and numeric sequences (e.g. version numbers, standalone digits, or multi‐digit numeric strings).

    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
    itorious
    -0.83
    cientes
    -0.82
     教程
    -0.81
    rament
    -0.81
     destacan
    -0.79
     esetén
    -0.76
    ortho
    -0.75
    aviar
    -0.73
    fea
    -0.73
    NOUNC
    -0.73
    POSITIVE LOGITS
    تمام
    0.79
    ialmente
    0.79
     buzzer
    0.77
     those
    0.77
     COS
    0.75
    宗
    0.75
    Jupiter
    0.74
    oplasty
    0.72
    ially
    0.72
     mattress
    0.71
    Activations Density 0.004%

    No Known Activations