© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 4058
    Prev
    Next
    INDEX
    Explanations

    code-related keywords such as "import", "except", and also operators commonly used in code like comma, dot, '!=' and "with"

    oai_token-act-pair · gemini-2.0-flash

    import

    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
     stesse
    -0.47
     pubblic
    -0.47
     vivent
    -0.45
     Deane
    -0.43
     dieux
    -0.43
     humaine
    -0.43
     chiens
    -0.43
     lå
    -0.41
     Amity
    -0.41
     hjel
    -0.39
    POSITIVE LOGITS
    AccessorTable
    0.86
    脚注の使い方
    0.82
     IBOutlet
    0.78
     CreateTagHelper
    0.76
    mybatisplus
    0.75
     MainAxisSize
    0.75
    awtextra
    0.75
     snippetHide
    0.72
    Obrázky
    0.69
    gypti
    0.69
    Activations Density 0.196%

    No Known Activations