© 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. 1743
    Prev
    Next
    INDEX
    Explanations

    words and phrases that indicate importance or subjective judgment

    oai_token-act-pair · gemini-2.0-flash

    important instructions or advice

    np_token-act-pair-logits · gpt-4o-mini
    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
    Erreferentziak
    -0.57
    rungsseite
    -0.55
     TextInputType
    -0.54
     ModelExpression
    -0.53
    WriteLiteral
    -0.53
    Datuak
    -0.52
    XmlAccessType
    -0.52
     BoxDecoration
    -0.51
     snippetHide
    -0.51
     NSCoder
    -0.51
    POSITIVE LOGITS
     to
    1.13
    ที่จะ
    0.57
     when
    0.57
     for
    0.51
    SpringBootTest
    0.51
    要把
    0.51
     abbraccio
    0.49
    Tikang
    0.49
     that
    0.49
     "];
    0.49
    Activations Density 7.761%

    No Known Activations