© 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. 19-GEMMASCOPE-TRANSCODER-16K
    7. 2912
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to negative outcomes, violence, death, or harm, often in the context of conflict or crime

    oai_token-act-pair · gemini-2.0-flash

    result or lead

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.19.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
     ipsum
    -0.46
    atella
    -0.46
    حيح
    -0.43
    getOut
    -0.43
    omin
    -0.40
    isic
    -0.40
    OrBuilder
    -0.39
    Page
    -0.38
    ttemberg
    -0.38
    classnames
    -0.38
    POSITIVE LOGITS
     resulted
    1.48
     lead
    1.37
    resulting
    1.27
    lead
    1.27
     leads
    1.26
     resulting
    1.23
     leading
    1.22
     result
    1.21
    leads
    1.20
     Leads
    1.18
    Activations Density 1.925%

    No Known Activations