© 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. 20-GEMMASCOPE-TRANSCODER-16K
    7. 11711
    Prev
    Next
    INDEX
    Explanations

    the word "said", often appearing in news articles and accompanied by a person's name or title, but also picks up on uses of the word "last."

    oai_token-act-pair · gemini-2.0-flash

    said

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.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
    깥
    -0.73
     estimés
    -0.71
     Signalez
    -0.62
    IndentedString
    -0.61
    出版年
    -0.60
    RegressionTest
    -0.58
     arşivlendi
    -0.58
     متحده
    -0.57
     utafiti
    -0.56
    adaptiveStyles
    -0.56
    POSITIVE LOGITS
     saying
    0.60
    saying
    0.51
    ChromeDriver
    0.50
     stating
    0.50
     yesterday
    0.46
    pointments
    0.44
     sayin
    0.41
     NKC
    0.41
    ்ச
    0.40
    omány
    0.40
    Activations Density 0.604%

    No Known Activations