© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 1783
    Prev
    Next
    INDEX
    Explanations

    the word "role", often preceded by words suggesting importance

    oai_token-act-pair · gemini-2.0-flash

    words indicating significance or roles

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

    role impact

    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
    Negative Logits
     in
    -0.57
     him
    -0.55
     nakalista
    -0.53
     nhau
    -0.51
    ТЬ
    -0.50
    ρός
    -0.50
    uzzi
    -0.50
    Искәрмәләр
    -0.48
     piece
    -0.46
    inot
    -0.46
    POSITIVE LOGITS
    rungsseite
    0.76
    WithIOException
    0.75
     that
    0.73
    NewUrlParser
    0.70
     quæ
    0.70
    #+#
    0.69
    HttpFoundation
    0.68
    SharedCtor
    0.67
     متعلقه
    0.66
     pleaſure
    0.66
    Activations Density 1.625%

    No Known Activations