© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 14939
    Prev
    Next
    INDEX
    Explanations

    narrative descriptions of jobs and people in a community

    oai_token-act-pair · gemini-2.0-flash

    job titles

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

    driver/server profession

    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
     église
    -0.55
     skolen
    -0.53
     solares
    -0.51
     liné
    -0.50
    mallows
    -0.49
     skrift
    -0.48
     isolada
    -0.48
     públicos
    -0.48
    HandlerContext
    -0.48
     vorstellen
    -0.47
    POSITIVE LOGITS
     driver
    1.34
     drivers
    1.14
    driver
    1.12
     Driver
    1.09
     DRIVER
    1.01
     Drivers
    0.98
    drivers
    0.98
     cook
    0.96
    Driver
    0.95
     waiter
    0.93
    Activations Density 3.487%

    No Known Activations