© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68429
    Prev
    Next
    INDEX
    Explanations

    statements about something

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

    The neuron selectively activates on numeric expressions (numbers, numeric modifiers, and tokens representing quantities or measurements).

    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
     verbess
    -1.92
    In
    -1.86
     reason
    -1.82
    <bos>
    -1.81
     Throughout
    -1.78
    ?”
    -1.75
     prevents
    -1.75
    !”
    -1.73
    ,”
    -1.73
     widely
    -1.71
    POSITIVE LOGITS
    
    1.96
    盉
    1.86
    ㅿ
    1.84
    婓
    1.78
    栳
    1.75
    のご紹介
    1.73
    ſſi
    1.71
    mability
    1.70
    genodigd
    1.68
    炻
    1.68
    Activations Density 0.015%

    No Known Activations