© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 1676
    Prev
    Next
    INDEX
    Explanations

    parts of words or names and possibly gene sequence letters

    oai_token-act-pair · gemini-2.0-flash

    patterns resembling names or titles

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

    non-English text

    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
    Negative Logits
     rospy
    -0.51
     canal
    -0.49
     campus
    -0.44
     scale
    -0.44
     pension
    -0.43
     Canal
    -0.43
     studio
    -0.43
     Mol
    -0.43
     seniors
    -0.43
    ysis
    -0.42
    POSITIVE LOGITS
     '\\;'
    0.93
    AsUp
    0.89
    expandindo
    0.79
    IntoConstraints
    0.71
     ligiloj
    0.66
     تضيفلها
    0.63
     stanovnika
    0.63
     disambiguazione
    0.62
    oredCriteria
    0.61
    PreferredItem
    0.60
    Activations Density 0.553%

    No Known Activations