© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5064
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "some" and the word "every", and phrases containing prepositions.

    oai_token-act-pair · gemini-2.0-flash

    variations of the word "where"

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

    where

    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
    Przypisy
    -0.57
     חיצוניים
    -0.56
     كومونز
    -0.56
    IMDG
    -0.56
    arschijnlijk
    -0.54
    Lähteet
    -0.54
    Inggris
    -0.54
    vgili
    -0.53
    ￡
    -0.53
    ollectionView
    -0.52
    POSITIVE LOGITS
     where
    2.77
    where
    2.75
    Where
    2.31
     Where
    2.14
    WHERE
    2.09
     WHERE
    2.00
     donde
    1.90
     où
    1.77
    donde
    1.60
     где
    1.57
    Activations Density 1.829%

    No Known Activations