© 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. 8981
    Prev
    Next
    INDEX
    Explanations

    code specifying the JSON format

    oai_token-act-pair · gemini-2.0-flash

    JSON content type headers

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

    content-type headers

    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
    SourceChecksum
    -0.54
    Fav
    -0.52
    anki
    -0.52
    #+#
    -0.51
     quæ
    -0.50
    ScopeManager
    -0.49
     pinulongan
    -0.49
    RectangleBorder
    -0.46
     TWA
    -0.46
     eas
    -0.46
    POSITIVE LOGITS
    TintMode
    0.68
     Audiodateien
    0.66
    
    0.65
    AsUp
    0.58
    Enlaces
    0.57
    Polizei
    0.56
     VIDEOTAPE
    0.55
    zeptember
    0.54
    AddTagHelper
    0.54
     DIVISION
    0.53
    Activations Density 0.650%

    No Known Activations