© 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. 12-GEMMASCOPE-TRANSCODER-16K
    7. 8306
    Prev
    Next
    INDEX
    Explanations

    This neuron appears to identify words or phrases within a broad range of categories, including personal pronouns, suffixes or word endings, legal terms, enzyme-related terms, disease characteristics and treatments, musical instruments, cultural terms,and positive descriptors.

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_12/width_16k/average_l0_6
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.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
    +#+#
    -0.89
     ligiloj
    -0.83
    writeFieldEnd
    -0.77
     autorytatywna
    -0.76
    SourceChecksum
    -0.75
    ⓧ
    -0.72
     QtGui
    -0.72
    WebElementEntity
    -0.71
     resourceCulture
    -0.69
    aarrggbb
    -0.69
    POSITIVE LOGITS
    <bos>
    0.66
     either
    0.35
    CompleteListener
    0.35
    ylene
    0.35
    反正
    0.34
     Orrell
    0.34
     Schie
    0.33
    philips
    0.32
    コー
    0.32
     not
    0.32
    Activations Density 7.653%

    No Known Activations