© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 11228
    Prev
    Next
    INDEX
    Explanations

    object instantiation or attribute access

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

    variable or object names immediately before method calls or property assignments in programming code.

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

    This neuron activates on tokens involved in setting or accessing object properties (e.g. lines assigning to prop["…"], emitter.position, mediaPlayer.Size, etc.).

    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
     or
    -2.06
     as
    -1.84
     this
    -1.77
     those
    -1.61
     akik
    -1.42
     alebo
    -1.36
     wspania
    -1.34
     empezaron
    -1.34
    หรือ
    -1.30
    >|</
    -1.28
    POSITIVE LOGITS
     also
    1.77
     גם
    1.74
     dimenti
    1.52
    詎
    1.46
    による
    1.45
    ：……
    1.45
     traduce
    1.39
     dirancang
    1.39
     tangga
    1.39
    erdings
    1.39
    Activations Density 0.013%

    No Known Activations