© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 11113
    Prev
    Next
    INDEX
    Explanations

    for loop in

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

    the keyword "in" when it appears in Python for-loop syntax.

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

    The neuron is primarily detecting numeric literals (number tokens) in the code.

    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
    ではでは
    -1.89
    にほんブログ村
    -1.66
    ではまた
    -1.64
    _
    -1.51
    ブログ村
    -1.51
    さてさて
    -1.46
    我们可以
    -1.45
    あー
    -1.43
    ごま
    -1.42
    鲩
    -1.42
    POSITIVE LOGITS
     are
    1.90
     \&
    1.72
     laikā
    1.53
     will
    1.49
     The
    1.45
    classnames
    1.39
     especially
    1.38
    ereien
    1.36
    女
    1.31
    議院
    1.30
    Activations Density 0.004%

    No Known Activations