Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 16-GEMMASCOPE-TRANSCODER-16K
    4. 16175
    Prev
    Next
    INDEX
    Explanations

    angle brackets, indentation, and table-like formatting

    oai_token-act-pair · gemini-2.0-flash

    code, symbols, and formatting

    np_token-act-pair-logits · gemini-2.0-flash

    non-natural language content

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_16/width_16k
    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
    Pick
    -0.44
    ric
    -0.41
    \
    -0.41
    The
    -0.41
     In
    -0.40
     One
    -0.40
     Pick
    -0.40
    One
    -0.40
    In
    -0.39
    խ
    -0.39
    POSITIVE LOGITS
     فريبيس
    1.10
    IndentedString
    0.91
     виправивши
    0.75
    principalColumn
    0.74
    
    0.73
     transfieras
    0.73
     ویکی‌پدی
    0.70
    LabelTagHelper
    0.69
     GoogleFonts
    0.68
    锈钢
    0.68
    Activations Density 1.213%

    No Known Activations