© 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. 15011
    Prev
    Next
    INDEX
    Explanations

    unicode characters and possibly some letter combinations

    oai_token-act-pair · gemini-2.0-flash

    foreign languages or names

    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
     E
    -0.48
     Hess
    -0.47
    vero
    -0.46
    raw
    -0.46
    wicz
    -0.46
     Gebrauch
    -0.45
    cję
    -0.45
     friends
    -0.44
     process
    -0.43
     AM
    -0.43
    POSITIVE LOGITS
    цездатний
    0.88
    tagext
    0.83
    writeFieldEnd
    0.80
     CascadeType
    0.73
     propOrder
    0.73
    ArrowToggle
    0.72
     WebDriverWait
    0.68
     CreateTagHelper
    0.68
    WriteBarrier
    0.65
    setVerticalGroup
    0.64
    Activations Density 0.533%

    No Known Activations