© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 6657
    Prev
    Next
    INDEX
    Explanations

    words that express value and support, but at different levels of activation.

    oai_token-act-pair · gemini-2.0-flash

    roles or professions

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

    jobs

    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
     consultato
    -0.65
    PerformLayout
    -0.63
    RenderAtEndOf
    -0.62
     craindre
    -0.61
     INTERESAR
    -0.58
     którzy
    -0.57
     uncles
    -0.56
    rungsseite
    -0.55
     eds
    -0.54
     toddlers
    -0.54
    POSITIVE LOGITS
     player
    1.31
     athlete
    1.17
     worker
    1.16
     investor
    1.14
     employee
    1.13
     developer
    1.12
     musician
    1.12
     artist
    1.12
     researcher
    1.10
     user
    1.09
    Activations Density 6.116%

    No Known Activations