© 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. 15544
    Prev
    Next
    INDEX
    Explanations

    instances of "&" or "Q&A" or other acronyms

    oai_token-act-pair · gemini-2.0-flash

    certain formatting patterns or special characters

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

    &

    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
     과
    -0.48
     Vogt
    -0.47
    autoload
    -0.47
    RESERV
    -0.46
     &
    -0.46
     du
    -0.45
    nicamente
    -0.45
     famí
    -0.45
     Gran
    -0.44
     Super
    -0.44
    POSITIVE LOGITS
     myſelf
    1.02
     itſelf
    0.86
     Reſ
    0.84
    ArgsConstructor
    0.82
     Anſ
    0.82
     iſt
    0.78
     whoſe
    0.77
     ſy
    0.77
     juſ
    0.76
     ་་
    0.76
    Activations Density 2.610%

    No Known Activations