© 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. 102851
    Prev
    Next
    INDEX
    Explanations

    license blocks

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

    metadata and configuration file syntax, particularly structured field names and attributes in code templates and data files.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron detects metadata field names and other header‐style tokens (e.g. version, tag, fileID, Instance, import, Scheme, BuildAction, etc.) that label sections of serialized or configuration data.

    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
     on
    -1.59
    gigi
    -1.57
     it
    -1.54
     in
    -1.50
     from
    -1.45
     any
    -1.45
    </h3>
    -1.38
     used
    -1.37
     should
    -1.36
     they
    -1.35
    POSITIVE LOGITS
     huile
    1.63
    Interested
    1.54
    licopter
    1.52
    those
    1.52
    umerable
    1.52
    omeness
    1.51
    SOME
    1.49
    狳
    1.46
    готовка
    1.46
    chables
    1.45
    Activations Density 0.031%

    No Known Activations