© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5487
    Prev
    Next
    INDEX
    Explanations

    code snippets particularly switch and case statements

    oai_token-act-pair · gemini-2.0-flash

    unique or specialized terms with unusual characters

    np_token-act-pair-logits · gpt-4o-mini

    say "case"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.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
     in
    -0.76
     I
    -0.63
     T
    -0.60
     
    -0.59
     n
    -0.57
     the
    -0.57
     x
    -0.56
     X
    -0.56
     '
    -0.55
     Mon
    -0.55
    POSITIVE LOGITS
     myſelf
    1.24
     itſelf
    1.20
     ―――――
    1.13
     ་་
    1.06
     Monfieur
    1.05
     pleaſure
    1.04
     houſe
    1.03
     Reſ
    1.02
     Houſe
    1.02
     Theſe
    1.02
    Activations Density 0.665%

    No Known Activations