© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 9990
    Prev
    Next
    INDEX
    Explanations

    capital letters related to medical or science documents, including names and abbreviations.

    oai_token-act-pair · gemini-2.0-flash

    Two capital letters

    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
    TagMode
    -1.09
     Efq
    -0.93
     itſelf
    -0.88
     Monfieur
    -0.88
     виправивши
    -0.88
     ſche
    -0.87
     Anſ
    -0.87
     raiſ
    -0.82
     myſelf
    -0.82
     houſe
    -0.81
    POSITIVE LOGITS
    ,
    0.75
    U
    0.60
    .
    0.60
    -
    0.58
    Z
    0.58
     -
    0.56
    Y
    0.55
    F
    0.54
    J
    0.54
    X
    0.52
    Activations Density 2.237%

    No Known Activations