© 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. 4-GEMMASCOPE-TRANSCODER-16K
    7. 1258
    Prev
    Next
    INDEX
    Explanations

    language contrasting present and past technological states, and describing the way things are done or methods used

    oai_token-act-pair · gemini-2.0-flash

    school, art, conventional

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
    abestanden
    -0.84
    RegistryLite
    -0.77
     AssemblyProduct
    -0.76
     myſelf
    -0.76
    licet
    -0.73
     itſelf
    -0.71
    setVerticalGroup
    -0.71
     juſt
    -0.69
    InputBorder
    -0.68
     oneself
    -0.68
    POSITIVE LOGITS
    Traditional
    0.60
    <bos>
    0.56
     traditional
    0.56
    traditional
    0.52
     clásica
    0.52
     classica
    0.50
     régi
    0.50
     classical
    0.50
     Traditional
    0.48
     conventional
    0.48
    Activations Density 3.115%

    No Known Activations