© 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. 15535
    Prev
    Next
    INDEX
    Explanations

    for loops with int i

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

    The neuron fires on numeric literals (integer constants) appearing in 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_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
     kabát
    -1.61
     histó
    -1.39
     vín
    -1.38
    ៨
    -1.34
     how
    -1.34
     Rumuni
    -1.23
    tambah
    -1.23
     Ayrıca
    -1.22
     kré
    -1.21
    הם
    -1.20
    POSITIVE LOGITS
     i
    1.61
     I
    1.52
    i
    1.38
    it
    1.30
    \
    1.26
    q
    1.25
    u
    1.23
    1.23
    w
    1.23
    ga
    1.19
    Activations Density 0.001%
    No Known Activations