© 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. 19-GEMMASCOPE-TRANSCODER-16K
    7. 5056
    Prev
    Next
    INDEX
    Explanations

    a variety of terms: Korean names, titles like manager, dates, delimiters e.g. parenthesis ), quotes ", and code snippets, but not in a consistent manner

    oai_token-act-pair · gemini-2.0-flash

    particularly

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.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
     صوتيه
    -0.54
     المعيارى
    -0.53
     reaſon
    -0.53
    chium
    -0.51
     Jefus
    -0.51
     ſtate
    -0.50
     juſ
    -0.49
     purpoſe
    -0.48
     Valentín
    -0.47
     Chriſt
    -0.46
    POSITIVE LOGITS
    ticularly
    0.68
     Particularly
    0.56
    InjectAttribute
    0.56
     RM
    0.56
    particularly
    0.55
     particuli
    0.54
     typelib
    0.54
     giovan
    0.53
     particularly
    0.52
     частности
    0.52
    Activations Density 1.067%

    No Known Activations