© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14082
    Prev
    Next
    INDEX
    Explanations

    code initialization and setup

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

    The neuron fires on code tokens related to initialization calls (e.g. “Initialize” in BeginInit/EndInit sequences).

    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
    鼉
    -1.73
     konfigura
    -1.51
    的で
    -1.47
     reklama
    -1.47
     wirklich
    -1.46
     quando
    -1.45
     gefeiert
    -1.45
     zape
    -1.45
     instruk
    -1.41
    蟊
    -1.41
    POSITIVE LOGITS
    <bos>
    1.59
    曲は
    1.48
     of
    1.44
    問題は
    1.38
    つまり
    1.34
    結局
    1.32
    は
    1.30
    愎
    1.28
    d
    1.28
    時間は
    1.28
    Activations Density 0.000%

    No Known Activations