© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 6743
    Prev
    Next
    INDEX
    Explanations

    text with specific formatting and abbreviations, possibly related to legal or technical documents including marking up text with bracket

    oai_token-act-pair · gemini-2.0-flash

    Unparsable text

    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
     '\\;'
    -0.68
     препратки
    -0.58
    <bos>
    -0.54
    urazione
    -0.53
     BorderRadius
    -0.52
     resultCode
    -0.50
    setObjectName
    -0.48
     King
    -0.48
    bulos
    -0.48
    IntoConstraints
    -0.48
    POSITIVE LOGITS
     تضيفلها
    0.65
    पया
    0.55
    RegistryLite
    0.50
     Inga
    0.50
    ()?;
    0.48
     مشين
    0.48
    ’.
    0.47
     NSCoder
    0.47
     gratu
    0.47
     Yol
    0.47
    Activations Density 1.146%

    No Known Activations