© 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. 9436
    Prev
    Next
    INDEX
    Explanations

    negative statements with contractions

    oai_token-act-pair · gemini-2.0-flash

    variations of the word 'one'

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

    saying "one"

    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
    WireFormatLite
    -0.67
     متعلقه
    -0.67
    Bioaccumulative
    -0.59
     للاسماء
    -0.56
    Vidite
    -0.56
     transfieras
    -0.56
    ftagPool
    -0.54
     незавершена
    -0.54
     SafeMath
    -0.52
    windowFixed
    -0.52
    POSITIVE LOGITS
     one
    0.94
     satu
    0.79
     één
    0.76
     ONE
    0.71
    Satu
    0.69
    satu
    0.61
     одному
    0.55
    shit
    0.55
     jednego
    0.55
    one
    0.55
    Activations Density 0.829%

    No Known Activations