© 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. 15220
    Prev
    Next
    INDEX
    Explanations

    quotes, particularly at the beginning or ends of phrases, and words like "said" that relate to quotes

    oai_token-act-pair · gemini-2.0-flash

    punctuation marks and sentence endings

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

    quotation marks

    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
    HTTPS
    -0.56
     gọn
    -0.48
    Portály
    -0.48
    pio
    -0.47
     Its
    -0.47
     ir
    -0.45
     meu
    -0.44
    Przeczytaj
    -0.43
    ṳ
    -0.43
    Cuál
    -0.43
    POSITIVE LOGITS
    AnchorStyles
    0.59
    enment
    0.56
    RegressionTest
    0.55
    NOPQRST
    0.54
     MenuView
    0.54
     createState
    0.52
    __":
    0.52
    nachron
    0.50
    encodeWith
    0.50
     []).
    0.50
    Activations Density 1.480%

    No Known Activations