© 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. 4-GEMMASCOPE-TRANSCODER-16K
    7. 1701
    Prev
    Next
    INDEX
    Explanations

    verbs with common suffixes or auxiliary verbs that relate to future or conditional events.

    oai_token-act-pair · gemini-2.0-flash

    Non-English and conversations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.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
     betweenstory
    -0.73
     Puch
    -0.56
    DoubleQuotes
    -0.56
    ViewImports
    -0.56
    lorette
    -0.55
    ริง
    -0.55
     secco
    -0.51
     AGB
    -0.51
    InSection
    -0.51
    تقاوى
    -0.51
    POSITIVE LOGITS
     will
    1.09
     Will
    0.96
    will
    0.90
    Will
    0.90
     WILL
    0.86
    <bos>
    0.78
    клопе
    0.72
    WILL
    0.71
     FetchType
    0.70
    المناصب
    0.70
    Activations Density 1.178%

    No Known Activations