© 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. 9906
    Prev
    Next
    INDEX
    Explanations

    the word 'while' and words used to connect clauses

    oai_token-act-pair · gemini-2.0-flash

    common phrases starting with 'while'

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

    while

    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
    Scénario
    -0.54
    =$((
    -0.50
    Kaynakça
    -0.49
     linkovi
    -0.49
     enquanto
    -0.49
     stället
    -0.49
    wahati
    -0.47
     laquelle
    -0.47
    #
    -0.47
     lequel
    -0.46
    POSITIVE LOGITS
     they
    0.88
     also
    0.79
     you
    0.73
     testifying
    0.71
     attempting
    0.69
    also
    0.68
     celebrating
    0.67
     the
    0.67
     contributing
    0.67
     displaying
    0.66
    Activations Density 3.347%

    No Known Activations