© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 3160
    Prev
    Next
    INDEX
    Explanations

    the word "upgrade" coupled with words about action, and possibly "hosting"

    oai_token-act-pair · gemini-2.0-flash

    upgrade

    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
    Negative Logits
     (
    -1.27
     
    -1.23
    s
    -1.19
     [
    -1.05
    ,
    -1.05
    <eos>
    -1.04
     I
    -1.02
     and
    -1.02
    ↵
    -0.99
    :
    -0.99
    POSITIVE LOGITS
     myſelf
    2.11
     ainfi
    1.99
     Efq
    1.98
     auffi
    1.92
     Monfieur
    1.89
     itſelf
    1.85
     pleaſure
    1.84
     avoient
    1.79
     themſelves
    1.79
     Theſe
    1.78
    Activations Density 1.585%

    No Known Activations