© 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. 4-GEMMASCOPE-TRANSCODER-16K
    7. 3328
    Prev
    Next
    INDEX
    Explanations

    words that express surprise, bewilderment, and exceptionality, and also references to entities with exceptional or supernatural abilities.

    oai_token-act-pair · gemini-2.0-flash

    weird/strange

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
     weird
    -1.49
     strange
    -1.48
     odd
    -1.37
     funny
    -1.27
    Strange
    -1.24
     wierd
    -1.20
     Strange
    -1.18
     Weird
    -1.18
    Odd
    -1.17
    Weird
    -1.17
    POSITIVE LOGITS
     cherchés
    0.66
    namic
    0.64
    numerusform
    0.57
    Personensuche
    0.56
     שוליים
    0.56
    aarrggbb
    0.55
    msgTypes
    0.52
    isContained
    0.52
     springfox
    0.51
    setVerticalGroup
    0.50
    Activations Density 8.159%

    No Known Activations