© 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-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 14985
    Prev
    Next
    INDEX
    Explanations

    code snippets involving drawing points, lines, or shapes in a graphics context

    oai_token-act-pair · gemini-2.0-flash

    code/mathematics

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    glTex
    -0.54
    ANTLR
    -0.52
    invokeLater
    -0.52
    sschutz
    -0.50
     '{@
    -0.49
     Lord
    -0.49
    ukone
    -0.49
    fir
    -0.48
    URLException
    -0.48
     memer
    -0.48
    POSITIVE LOGITS
     cdti
    0.56
    AsUp
    0.53
     noqa
    0.52
    TagNumber
    0.50
    AddTagHelper
    0.50
    AndEndTag
    0.50
    ksiä
    0.50
    Tikang
    0.48
     Rho
    0.48
    TargetException
    0.48
    Activations Density 1.435%

    No Known Activations