© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 5880
    Prev
    Next
    INDEX
    Explanations

    the beginning of the word "suppose" or "superior" or "supreme"

    oai_token-act-pair · gemini-2.0-flash

    sup

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     flat
    -0.53
    <bos>
    -0.52
    cloudfront
    -0.51
     S
    -0.50
     Hal
    -0.50
     Sol
    -0.50
    ده
    -0.49
     sou
    -0.49
     hal
    -0.49
    作
    -0.49
    POSITIVE LOGITS
     autorytatywna
    0.68
    writeFieldEnd
    0.67
    脚注の使い方
    0.67
     nationaux
    0.66
     HasFactory
    0.66
     transfieras
    0.65
     lenker
    0.63
     kaarangay
    0.59
     Мексичка
    0.59
     étoient
    0.58
    Activations Density 0.513%

    No Known Activations