© 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. 7030
    Prev
    Next
    INDEX
    Explanations

    statistical data and formulas, especially those containing p-values and inequalities

    oai_token-act-pair · gemini-2.0-flash

    numeric values and statistical comparisons

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

    less than symbol

    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
    addCriterion
    -1.00
     propOrder
    -0.93
    CppMethod
    -0.91
     autorytatywna
    -0.87
     &___
    -0.87
    __":
    -0.87
    исленность
    -0.85
     ویکی‌پدیا
    -0.84
     насељу
    -0.84
    tagHelperRunner
    -0.81
    POSITIVE LOGITS
     
    0.43
     **
    0.38
     loss
    0.38
    shad
    0.38
     eighth
    0.37
     eight
    0.37
    rena
    0.37
     fall
    0.36
    0.36
     ");
    0.35
    Activations Density 6.823%

    No Known Activations