© 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. 22-GEMMASCOPE-RES-131K
    4. 64982
    Prev
    Next
    INDEX
    Explanations

    programming constructs and names

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

    The neuron fires on programming API identifiers—especially method, class, or framework call names (e.g. CreateElement, AppendChild, PdfPTable, DefaultIfEmpty).

    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
     and
    -1.66
     diesem
    -1.46
    _
    -1.42
    A
    -1.40
    になったら
    -1.39
     ¥
    -1.36
     this
    -1.33
    For
    -1.33
    At
    -1.32
    "/"
    -1.28
    POSITIVE LOGITS
     (
    1.80
    では
    1.66
    यह
    1.57
    方針
    1.55
    دانلود
    1.52
    <bos>
    1.51
    TRIBUN
    1.48
     використовувати
    1.45
    																				
    1.44
     gelatina
    1.41
    Activations Density 0.021%

    No Known Activations