© 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. 20-GEMMASCOPE-TRANSCODER-16K
    7. 13010
    Prev
    Next
    INDEX
    Explanations

    names of people in a story, particularly when a possessive is used

    oai_token-act-pair · gemini-2.0-flash

    Characters in narratives

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.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
     would
    -0.61
    would
    -0.56
     akan
    -0.55
     will
    -0.54
    Would
    -0.54
     zou
    -0.53
    shall
    -0.52
     Would
    -0.51
     should
    -0.51
     shall
    -0.50
    POSITIVE LOGITS
     itſelf
    0.79
    RenderAtEndOf
    0.77
     ་་
    0.74
     ―――――
    0.71
    annica
    0.69
     leaſt
    0.69
     Cæsar
    0.68
     myſelf
    0.66
    >\<^
    0.66
    $.
    0.66
    Activations Density 1.436%

    No Known Activations