© 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. 12629
    Prev
    Next
    INDEX
    Explanations

    mentions of percentages, numeric data, and related words.

    oai_token-act-pair · gemini-2.0-flash

    say "percentage"

    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
    CodedInputStream
    -0.70
     дописавши
    -0.69
    ])));
    -0.67
    ;");
    -0.65
    сюда
    -0.63
    /");
    -0.61
    ]";
    -0.61
    liefert
    -0.61
     blessures
    -0.61
     oprot
    -0.60
    POSITIVE LOGITS
    %
    0.59
    WithIOException
    0.57
    broek
    0.52
     per
    0.48
     p
    0.46
     par
    0.45
    autorest
    0.44
    multirow
    0.44
     percent
    0.43
     satisfied
    0.43
    Activations Density 8.922%

    No Known Activations