© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 2424
    Prev
    Next
    INDEX
    Explanations

    text with a lot of numbers and references to sections or rules

    oai_token-act-pair · gemini-2.0-flash

    technical language and jargon

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

    3

    np_max-act-logits · gemini-2.0-flash

    numbers

    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_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
     AssemblyCulture
    -1.11
    FormTagHelper
    -0.80
    StoryboardSegue
    -0.75
    RenderAtEndOf
    -0.67
     незавершена
    -0.64
    MigrationBuilder
    -0.61
    ruvate
    -0.60
    íncia
    -0.55
    IVEREF
    -0.54
    WriteBarrier
    -0.54
    POSITIVE LOGITS
    PerformLayout
    0.66
     Pingback
    0.52
    twimg
    0.46
    irov
    0.44
     measuring
    0.43
    msgTypes
    0.42
    timing
    0.42
    BagLayout
    0.42
    ✨:
    0.42
     обращения
    0.41
    Activations Density 41.982%

    No Known Activations