© 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. 7144
    Prev
    Next
    INDEX
    Explanations

    code with an indent of two spaces

    oai_token-act-pair · gemini-2.0-flash

    special characters or unique patterns

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

    C

    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
     
    -0.78
     le
    -0.73
     F
    -0.65
     B
    -0.64
     in
    -0.64
     (
    -0.63
     i
    -0.63
    <eos>
    -0.62
     D
    -0.61
     I
    -0.60
    POSITIVE LOGITS
     itſelf
    1.23
     myſelf
    1.17
     ―――――
    1.14
     Monfieur
    1.11
     Jefus
    1.09
     Eſ
    1.04
     becauſe
    1.04
     Efq
    1.03
     pleaſure
    1.03
     Reſ
    1.02
    Activations Density 0.369%

    No Known Activations