© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13653
    Prev
    Next
    INDEX
    Explanations

    url, Q, Do, Field

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

    This neuron detects numeric tokens (especially numerical literals with digits and decimals).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    <bos>
    -2.17
    
    -2.03
    ↵↵↵↵↵↵↵↵↵↵↵↵↵
    -1.91
     donker
    -1.88
    沵
    -1.85
     nevera
    -1.84
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -1.82
     eiland
    -1.82
    橚
    -1.81
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    -1.79
    POSITIVE LOGITS
     you
    2.55
    when
    1.85
    側は
    1.65
     destacan
    1.62
    people
    1.56
     ингреди
    1.53
    hnout
    1.52
    瀟
    1.52
     navigate
    1.48
    了自己的
    1.48
    Activations Density 0.000%

    No Known Activations