© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 9923
    Prev
    Next
    INDEX
    Explanations

    the word "while" alongside other seemingly random words (e.g. "hardware", "who", "content", "unlike")

    oai_token-act-pair · gemini-2.0-flash

    while

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
     while
    -1.00
     
    -0.92
      
    -0.86
     when
    -0.83
     whilst
    -0.77
    <eos>
    -0.77
     in
    -0.75
     for
    -0.75
     although
    -0.74
     with
    -0.72
    POSITIVE LOGITS
     themſelves
    1.38
     itſelf
    1.34
     myſelf
    1.33
    ſelves
    1.29
     uſed
    1.28
     auffi
    1.27
    ſelf
    1.27
     purpoſe
    1.24
     ―――――
    1.23
     ſta
    1.23
    Activations Density 1.043%

    No Known Activations