© 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. 4410
    Prev
    Next
    INDEX
    Explanations

    negative statements with the word "widen" or references to something covering a lot such as in "Sandy is like spackle: it covers a whole lot of mistakes"

    oai_token-act-pair · gemini-2.0-flash

    large size

    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
    DataAnnotations
    -0.65
    BeginContext
    -0.63
    ViewImports
    -0.61
     Chwiliwch
    -0.61
     DbSet
    -0.59
     VizieR
    -0.58
    انيف
    -0.57
    htbp
    -0.54
    Хьажоргаш
    -0.54
    ंदीखरीदारी
    -0.54
    POSITIVE LOGITS
     deep
    0.97
     huge
    0.95
    huge
    0.93
    Huge
    0.85
     Huge
    0.79
     large
    0.77
     enormous
    0.77
     HUGE
    0.77
     gigantic
    0.76
     big
    0.75
    Activations Density 2.134%

    No Known Activations