© 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. 84174
    Prev
    Next
    INDEX
    Explanations

    years, money, and dates

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

    The neuron fires on numeric tokens, especially multi‐digit and decimal number sequences (measurements or statistics).

    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
     امنیت
    -0.74
    larıyla
    -0.73
    voet
    -0.72
    ract
    -0.72
     douane
    -0.71
    ounder
    -0.71
     aseguró
    -0.70
     Mistakes
    -0.69
    coffee
    -0.69
     folder
    -0.69
    POSITIVE LOGITS
     fear
    0.81
    fear
    0.69
     brainstorm
    0.68
    遮
    0.68
    sns
    0.67
    camiento
    0.67
    gerea
    0.66
     Carter
    0.65
     visión
    0.65
    llywood
    0.65
    Activations Density 0.047%

    No Known Activations