© 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. 13510
    Prev
    Next
    INDEX
    Explanations

    positive descriptive terms and words referencing duration or clarity

    oai_token-act-pair · gemini-2.0-flash

    Describing quality or states

    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
    <bos>
    -0.77
     to
    -0.68
     that
    -0.60
     itself
    -0.56
     for
    -0.55
     as
    -0.54
     but
    -0.54
     something
    -0.48
     because
    -0.48
     enough
    -0.47
    POSITIVE LOGITS
    HomeAsUpEnabled
    0.99
    tvguidetime
    0.89
     Monfieur
    0.84
    adecimal
    0.79
    DispatchToProps
    0.77
    Espèce
    0.76
     luxurious
    0.75
    XSSF
    0.75
    onViewCreated
    0.75
    AndEndTag
    0.74
    Activations Density 1.626%

    No Known Activations