© 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. 3944
    Prev
    Next
    INDEX
    Explanations

    words relating to fantasy or science fiction concepts involving magic, mythical creatures, or supernatural powers

    oai_token-act-pair · gemini-2.0-flash

    magic and fantasy

    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
     الحره
    -0.59
     uLocal
    -0.52
     kiệm
    -0.51
    onAttach
    -0.50
     للمعارف
    -0.49
     régl
    -0.47
     paura
    -0.47
    ingredients
    -0.47
    obraz
    -0.47
    <thead>
    -0.47
    POSITIVE LOGITS
    Tembelea
    0.68
    #+#
    0.63
    mark
    0.61
     mark
    0.58
     Serikat
    0.56
     disambiguazione
    0.54
    marks
    0.52
     TType
    0.50
    keyp
    0.50
    jsonwebtoken
    0.50
    Activations Density 2.250%

    No Known Activations