© 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. 2919
    Prev
    Next
    INDEX
    Explanations

    uses of the word "thing", references to Facebook, and JSON formatting

    oai_token-act-pair · gemini-2.0-flash

    patterns indicating variable names or JSON properties

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

    letter X

    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
     Jefus
    -1.14
     Theſe
    -1.02
     raiſ
    -1.00
    ImageContext
    -0.97
     becauſe
    -0.93
     Beſ
    -0.93
     ſtand
    -0.92
     itſelf
    -0.92
     Reſ
    -0.90
     houſe
    -0.89
    POSITIVE LOGITS
     X
    0.73
    X
    0.59
     __
    0.57
     [
    0.56
     x
    0.56
     (
    0.50
    __
    0.49
    x
    0.47
     millioner
    0.45
    üte
    0.43
    Activations Density 3.635%

    No Known Activations