© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 100824
    Prev
    Next
    INDEX
    Explanations

    string formatting and variables

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

    string formatting syntax in programming code, particularly format strings with placeholders.

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

    The neuron primarily detects numeric literals (numbers, including integers and floats) 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
    不明白
    -1.52
     إذا
    -1.46
     этих
    -1.44
    
    -1.44
    也很
    -1.41
    dico
    -1.40
    *{-
    -1.38
     asegurarse
    -1.37
     maleta
    -1.35
     averiguar
    -1.34
    POSITIVE LOGITS
     the
    1.91
     our
    1.66
    ).
    1.49
     this
    1.48
     those
    1.45
     your
    1.35
     their
    1.33
     видел
    1.29
    ~、
    1.21
     which
    1.20
    Activations Density 0.006%

    No Known Activations