© 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. 3336
    Prev
    Next
    INDEX
    Explanations

    places, especially in Asia and Eastern Europe, as well as words that often come before or after places, such as "found in".

    oai_token-act-pair · gemini-2.0-flash

    technical jargon related to web or software functionality

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

    Locations and regions

    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
    Kariera
    -0.55
     samband
    -0.48
    portál
    -0.47
     Hearn
    -0.43
     ujednoznacz
    -0.43
     valiente
    -0.43
    Kalb
    -0.43
    ANNES
    -0.43
     aniversario
    -0.42
    idão
    -0.42
    POSITIVE LOGITS
    uxxxx
    0.69
    UnsafeEnabled
    0.65
     itſelf
    0.60
    consult
    0.60
    日閲覧
    0.58
     consult
    0.58
    UrlResolution
    0.57
    SharedCtor
    0.56
    Consult
    0.56
    ThroughAttribute
    0.56
    Activations Density 1.082%

    No Known Activations