© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 10557
    Prev
    Next
    INDEX
    Explanations

    lines from a rap song or poem

    oai_token-act-pair · gemini-2.0-flash

    unconventional or unique words

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

    rambling expressive 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
    Not in Any Lists

    No Comments

    Negative Logits
     
    -0.56
     alternativ
    -0.50
     which
    -0.47
     nov
    -0.45
     d
    -0.45
     [
    -0.45
     Mandel
    -0.45
    <bos>
    -0.45
     here
    -0.44
     mon
    -0.44
    POSITIVE LOGITS
    TemporalType
    0.79
    IVEREF
    0.72
     Anſ
    0.72
     متعلقه
    0.71
     varandra
    0.71
    Tikang
    0.70
    invokeLater
    0.70
     ―――――
    0.70
     myſelf
    0.69
     vulgaires
    0.69
    Activations Density 3.702%

    No Known Activations