© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67440
    Prev
    Next
    INDEX
    Explanations

    Paget disease, Pagophilus, Pagano, pagers

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

    This neuron fires on occurrences of the token “Pag” (as in page references or pagination markers).

    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
     imprimi
    -2.67
     precisamos
    -2.64
    丏
    -2.61
     และ
    -2.55
    лением
    -2.50
     grabar
    -2.44
     dieß
    -2.39
    傒
    -2.38
    仃
    -2.36
    
    -2.33
    POSITIVE LOGITS
    e
    2.95
    un
    2.70
    ag
    2.61
    的
    2.61
    <bos>
    2.59
     enticing
    2.55
     unwavering
    2.53
    p
    2.53
    d
    2.53
    ')
    2.52
    Activations Density 0.008%

    No Known Activations