© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 47977
    Prev
    Next
    INDEX
    Explanations

    cryptography, declaration, Japanese

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

    The neuron strongly activates on numeric literal tokens (digits and multi-digit numbers) in code or 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
     paradas
    -0.78
    Dettagli
    -0.78
     fortale
    -0.78
    န်
    -0.77
    Dimanche
    -0.76
    Cela
    -0.75
    Année
    -0.73
     ^{-
    -0.73
     космети
    -0.71
    selectBy
    -0.71
    POSITIVE LOGITS
    6
    0.90
    FileVersion
    0.87
    5
    0.78
    8
    0.75
    paramInt
    0.73
    DECLARATION
    0.72
    4
    0.72
     japonaise
    0.71
    Declaration
    0.70
     kanji
    0.69
    Activations Density 0.038%

    No Known Activations