© 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. 15908
    Prev
    Next
    INDEX
    Explanations

    words from a language other than English, likely Croatian

    oai_token-act-pair · gemini-2.0-flash

    Non-English code 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
    ngdoc
    -0.94
    rungsseite
    -0.88
     JpaRepository
    -0.87
     disambiguazione
    -0.83
     للاسماء
    -0.81
     Majefty
    -0.78
    RTDA
    -0.77
     ſeveral
    -0.76
     ddelweddau
    -0.75
    tagHelperRunner
    -0.74
    POSITIVE LOGITS
     such
    0.45
    such
    0.39
     Sche
    0.38
    ero
    0.37
     Such
    0.36
    one
    0.35
     Ro
    0.34
    chränk
    0.34
     The
    0.34
    tree
    0.34
    Activations Density 0.142%

    No Known Activations