© 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. 22-GEMMASCOPE-RES-131K
    7. 15095
    Prev
    Next
    INDEX
    Explanations

    cell text Self Q

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

    The neuron fires on programming-style identifiers and method calls (e.g. class names like PdfPTable or PdfPCell and calls like doc.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_22/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
     who
    -1.38
     to
    -1.34
     akku
    -1.23
     how
    -1.23
     sakin
    -1.15
     Meksi
    -1.13
    viale
    -1.12
     kuid
    -1.09
     刷
    -1.09
    ところに
    -1.09
    POSITIVE LOGITS
    暧
    1.16
    ATE
    1.16
    okhttp
    1.12
    ಔ
    1.12
     داشتند
    1.10
    Actu
    1.09
     calificación
    1.09
     hark
    1.09
    ӥ
    1.09
    atorio
    1.08
    Activations Density 0.000%

    No Known Activations