© 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. 2595
    Prev
    Next
    INDEX
    Explanations

    html tags, especially `<script>`, `<link>`, `<body>`, `<head>`, `<a>`, and `<div>`

    oai_token-act-pair · gemini-2.0-flash

    HTML/XML elements and attributes

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

    Code or HTML

    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
    principalColumn
    -0.66
     pulumi
    -0.57
    msgSender
    -0.55
    Tikang
    -0.51
    apimachinery
    -0.50
    putString
    -0.49
    esez
    -0.46
     désolés
    -0.46
    StoryboardSegue
    -0.46
    例句
    -0.46
    POSITIVE LOGITS
    Personendaten
    0.80
     <
    0.68
    AnimationsModule
    0.63
    <
    0.60
    <%=
    0.56
    קישורים
    0.56
     edit
    0.53
     Edit
    0.53
     thereof
    0.53
     (<
    0.52
    Activations Density 1.004%

    No Known Activations