© 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. 93746
    Prev
    Next
    INDEX
    Explanations

    without further ado

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

    The neuron fires on sequences of digits (and adjacent punctuation or letters) identifying things like bond or stock identifiers, dates, license versions, 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_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
    lotten
    -0.91
    Knowledge
    -0.77
     Knowledge
    -0.73
    Coffee
    -0.73
     schnell
    -0.70
    عل
    -0.70
    PropertyValue
    -0.70
    Blue
    -0.68
    volumen
    -0.68
    衿
    -0.68
    POSITIVE LOGITS
     sooth
    0.71
    עו
    0.71
     孩子
    0.71
     Rond
    0.70
    rmızı
    0.69
    ホー
    0.69
    するために
    0.66
     interactive
    0.65
     امنیت
    0.65
     Extre
    0.65
    Activations Density 0.074%

    No Known Activations