© 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. 111850
    Prev
    Next
    INDEX
    Explanations

    lower and upper limits

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

    The neuron detects numeric values that specify upper or lower bounds (e.g. “upper bound,” “lower limit,” 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_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
     enige
    -1.28
    茘
    -1.20
     reconna
    -1.14
     The
    -1.12
     privilegi
    -1.11
     habitudes
    -1.11
    
    -1.11
    attitude
    -1.09
    Ꮈ
    -1.09
     رمضان
    -1.08
    POSITIVE LOGITS
    上限
    1.34
    breviations
    1.33
    إذا
    1.32
     excepción
    1.28
    ギリス
    1.28
     excitedly
    1.27
    例外
    1.24
    のための
    1.22
    𓄹
    1.20
    学科
    1.19
    Activations Density 0.034%

    No Known Activations