© 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-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 9448
    Prev
    Next
    INDEX
    Explanations

    numerical identifiers often used in technical documentation or lists

    oai_token-act-pair · gemini-2.0-flash

    time durations and quantities

    np_token-act-pair-logits · gpt-4o-mini

    2

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    queryInterface
    -0.74
    MigrationBuilder
    -0.73
    awaiter
    -0.71
    unzel
    -0.69
    DockStyle
    -0.66
    ChildScrollView
    -0.66
    fillType
    -0.65
    ArrowToggle
    -0.63
     виправивши
    -0.63
    QPalette
    -0.63
    POSITIVE LOGITS
     minute
    0.67
     bedrooms
    0.65
     bedroom
    0.64
     years
    0.61
     year
    0.56
     weeks
    0.54
     months
    0.53
     hours
    0.52
     days
    0.51
     Bedroom
    0.49
    Activations Density 3.005%

    No Known Activations