© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 40105
    Prev
    Next
    INDEX
    Explanations

    upper and lower bounds

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

    This neuron detects numeric “bound” expressions—particularly instances of upper or lower bounds (and their accompanying numerical values) in mathematical 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_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
    )}=\
    -1.79
     sortes
    -1.73
    🧌
    -1.72
    )$\\
    -1.67
    lahraga
    -1.66
    ’
    -1.66
    %")
    -1.65
     新款
    -1.65
    
    -1.65
     aperta
    -1.63
    POSITIVE LOGITS
     of
    2.67
    er
    2.06
    up
    1.79
    s
    1.77
     سایر
    1.76
    他の
    1.74
     worldwide
    1.70
    
    1.68
     others
    1.64
     the
    1.61
    Activations Density 0.010%

    No Known Activations