© 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. 10-GEMMASCOPE-RES-131K
    4. 103004
    Prev
    Next
    INDEX
    Explanations

    object, value, properties, data, state

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

    function or method names in code, particularly those related to retrieving, getting, or accessing values and properties.

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

    The neuron activates on code identifiers and comments containing “prop” or “value(s),” i.e. tokens referring to properties and their values.

    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
     должны
    -1.46
     llegue
    -1.27
    蟳
    -1.23
     после
    -1.21
     will
    -1.16
    אם
    -1.16
    satunya
    -1.15
    would
    -1.15
    ビット
    -1.15
    uert
    -1.13
    POSITIVE LOGITS
     of
    2.06
     before
    1.76
     with
    1.74
     through
    1.52
     while
    1.47
     as
    1.44
     for
    1.37
     using
    1.34
    …]
    1.24
     when
    1.23
    Activations Density 0.038%

    No Known Activations