© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 20-GEMMASCOPE-TRANSCODER-16K
    7. 15461
    Prev
    Next
    INDEX
    Explanations

    words prefixed with "non", "uns", and "bio" and words ending in "ings"

    oai_token-act-pair · gemini-2.0-flash

    scientific terminology

    np_max-act-logits · gemini-2.0-flash

    word prefixes Method used: 1. Reason: tokens are beginnings of longer words (e.g., Uns- → Unsatisfactory, immunos- → immunostimulatory)

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.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
    ISODE
    -0.60
    complexContent
    -0.54
     Económica
    -0.54
     înc
    -0.50
    sió
    -0.49
    shund
    -0.49
     sherds
    -0.48
    ocarcinoma
    -0.48
    leetcode
    -0.47
     vellum
    -0.47
    POSITIVE LOGITS
    InSection
    0.69
    (!__
    0.63
    NameInMap
    0.57
     nakalista
    0.57
     EconPapers
    0.55
     ExecuteAsync
    0.54
    WithIOException
    0.54
    +#+#
    0.54
    DoubleQuotes
    0.51
    UnsafeEnabled
    0.50
    Activations Density 1.044%

    No Known Activations