© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 26935
    Prev
    Next
    INDEX
    Explanations

    San Francisco, I DON'T, Total intracranial, Tissue culture

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

    The neuron activates on import statements that reference major framework “core” modules (e.g. '@angular/core') or library names like “react,” i.e. when a code import sources a principal framework package.

    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
     from
    -1.71
     to
    -1.63
     for
    -1.55
     each
    -1.46
     by
    -1.46
     all
    -1.27
     against
    -1.22
     throughout
    -1.16
     between
    -1.13
     different
    -1.11
    POSITIVE LOGITS
    donde
    1.20
    がありますが
    1.19
    Rumah
    1.12
    Tempat
    1.11
    mientras
    1.09
     aras
    1.09
    ᄇ
    1.08
    verstanden
    1.07
    Bjs
    1.06
    Recept
    1.06
    Activations Density 0.000%

    No Known Activations