© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 7048
    Prev
    Next
    INDEX
    Explanations

    words and symbols from computer code, email headers, and question/answer formats

    oai_token-act-pair · gemini-2.0-flash

    words with archaic or stylized spelling

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

    say "preposition or uppercase word"

    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
     de
    -0.57
     di
    -0.57
     et
    -0.56
     al
    -0.56
     in
    -0.54
    ,
    -0.54
     un
    -0.53
     that
    -0.52
     on
    -0.52
     my
    -0.51
    POSITIVE LOGITS
     Theſe
    1.27
     myſelf
    1.20
     Jefus
    1.16
     Anſ
    1.16
    ſelves
    1.13
     Monfieur
    1.13
    ſelf
    1.11
     itſelf
    1.08
     faſt
    1.05
     Beſ
    1.04
    Activations Density 3.383%

    No Known Activations