© 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. 22-GEMMASCOPE-TRANSCODER-16K
    7. 4859
    Prev
    Next
    INDEX
    Explanations

    complicated math, chemical formulas, code snippets, and figure references

    oai_token-act-pair · gemini-2.0-flash

    Code/math snippets

    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
     lib
    -0.65
     Mase
    -0.63
    locks
    -0.60
    mase
    -0.58
    lizard
    -0.57
     locks
    -0.57
    lib
    -0.57
    lius
    -0.57
     lock
    -0.57
    locker
    -0.57
    POSITIVE LOGITS
     mergeFrom
    0.61
    +#+#
    0.61
     /\.(
    0.56
     насељу
    0.55
    AndEndTag
    0.54
     propOrder
    0.54
     AssemblyCulture
    0.53
     esternos
    0.53
     atentos
    0.52
     vulgaires
    0.52
    Activations Density 0.771%

    No Known Activations