© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 2773
    Prev
    Next
    INDEX
    Explanations

    both parts of the document relating to formatted code in HTML and Chinese text, along with some tokens from the HTML

    oai_token-act-pair · gemini-2.0-flash

    technical text/code

    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
    Negative Logits
     whoſe
    -0.77
     >=",
    -0.77
     Efq
    -0.76
     their
    -0.74
    their
    -0.74
    ]]);
    -0.72
    []=$
    -0.71
    SourceChecksum
    -0.70
     Италијани
    -0.70
     becauſe
    -0.69
    POSITIVE LOGITS
    出版年
    0.60
    期
    0.40
    津
    0.39
    ลิต
    0.37
    CppMethod
    0.36
     blan
    0.35
     createContext
    0.35
    bes
    0.34
     mied
    0.34
     đãi
    0.34
    Activations Density 0.250%

    No Known Activations