© 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. 15974
    Prev
    Next
    INDEX
    Explanations

    words that are commonly used to introduce clauses

    oai_token-act-pair · gemini-2.0-flash

    where

    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
    how
    -0.59
    How
    -0.56
     How
    -0.52
    I
    -0.48
    What
    -0.47
    which
    -0.44
     What
    -0.43
    HOW
    -0.43
     &
    -0.41
    Linq
    -0.41
    POSITIVE LOGITS
     ſhe
    1.32
     they
    1.30
     ſeveral
    1.30
     Monfieur
    1.27
     itſelf
    1.27
     myſelf
    1.23
     Efq
    1.21
     Jefus
    1.20
     becauſe
    1.16
     we
    1.15
    Activations Density 20.219%

    No Known Activations