© 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. 14794
    Prev
    Next
    INDEX
    Explanations

    printf and format specifiers

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

    The neuron fires on printf‐/scanf‐style format specifiers (the “%” tokens and their following type letters).

    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
    がなく
    -1.73
    を
    -1.72
    などが
    -1.55
    CCIÓN
    -1.54
    がある
    -1.51
    UCIÓN
    -1.49
    がありますが
    -1.49
     be
    -1.49
    などの
    -1.48
    <bos>
    -1.44
    POSITIVE LOGITS
    為に
    2.14
    のだろう
    1.73
    ほら
    1.67
    向けの
    1.59
    ようになりました
    1.58
    事です
    1.57
    明日は
    1.55
    ようになる
    1.55
    ようになります
    1.52
    場所で
    1.52
    Activations Density 0.004%

    No Known Activations