© 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. 34-GEMMASCOPE-RES-131K
    7. 78274
    Prev
    Next
    INDEX
    Explanations

    class instantiation and extension

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

    The neuron strongly activates on occurrences of the PHP “new” keyword when instantiating a class or object.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    场所
    -0.86
     IService
    -0.83
    甚至
    -0.80
     Immunology
    -0.80
    worms
    -0.80
     кількість
    -0.78
    lorette
    -0.78
     большую
    -0.77
    stino
    -0.76
     сложности
    -0.75
    POSITIVE LOGITS
     stdClass
    0.96
    CI
    0.89
     extends
    0.89
     Верхов
    0.85
    Iden
    0.84
    std
    0.82
    дравству
    0.81
     remains
    0.81
     CI
    0.80
    classname
    0.79
    Activations Density 0.031%

    No Known Activations