© 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. 22-GEMMASCOPE-RES-131K
    7. 65469
    Prev
    Next
    INDEX
    Explanations

    piecewise constructor

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

    This neuron detects code tokens where a variable or identifier is being assigned a numeric literal (i.e. “identifier = number”).

    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
    〜。
    -1.44
    与此同时
    -1.39
    Possibly
    -1.38
     ideen
    -1.38
    そして
    -1.38
    Apparently
    -1.34
    その
    -1.33
    ～。
    -1.31
    Only
    -1.29
     kupić
    -1.29
    POSITIVE LOGITS
    の準備
    1.42
    んでしょう
    1.36
    の方に
    1.33
    聴いて
    1.30
    んですか
    1.29
    えません
    1.28
    参考に
    1.25
     from
    1.24
    側から
    1.23
    の人に
    1.23
    Activations Density 0.000%

    No Known Activations