© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 16172
    Prev
    Next
    INDEX
    Explanations

    reports, analysis, and government bodies

    oai_token-act-pair · gemini-2.0-flash

    report

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
    ena
    -0.38
    Py
    -0.37
    ra
    -0.37
    node
    -0.37
    da
    -0.37
    лене
    -0.36
    ED
    -0.35
    ts
    -0.35
    őd
    -0.35
    tri
    -0.35
    POSITIVE LOGITS
     report
    2.16
    report
    1.87
     Report
    1.80
     reports
    1.79
     REPORT
    1.75
    Report
    1.67
     Reports
    1.61
     REPORTS
    1.54
    reports
    1.52
    REPORT
    1.49
    Activations Density 2.603%

    No Known Activations