© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 8-GEMMASCOPE-TRANSCODER-16K
    7. 2471
    Prev
    Next
    INDEX
    Explanations

    words and phrases common in scientific and legal writing such as 'were', 'agents that were', 'result is made after', and parenthetical citations

    oai_token-act-pair · gemini-2.0-flash

    scientific/legal texts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.8.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
    ChildIndex
    -0.75
    脚注の使い方
    -0.64
    homonymie
    -0.58
    äiv
    -0.54
     وتسجيلات
    -0.52
     "..\..\..\
    -0.50
    enschappelijke
    -0.48
    crity
    -0.47
     transférez
    -0.46
    ścic
    -0.46
    POSITIVE LOGITS
     ujednoznacz
    0.54
    Eloquent
    0.53
    Lö
    0.52
    ImageContext
    0.52
    arsal
    0.50
    ⓘ
    0.50
    ื่อง
    0.48
     resourceCulture
    0.48
    anthes
    0.48
    rungsseite
    0.47
    Activations Density 19.003%

    No Known Activations