© 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. 6812
    Prev
    Next
    INDEX
    Explanations

    the word "Win" or "win", sometimes followed by related characters

    oai_token-act-pair · gemini-2.0-flash

    the word "Win"

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

    win

    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
    FragmentManager
    -0.52
    Disposed
    -0.50
    SpringBootTest
    -0.50
    Архівовано
    -0.48
    нато
    -0.48
     $__
    -0.47
    writeField
    -0.46
    brainly
    -0.46
    ragan
    -0.45
     PyTuple
    -0.45
    POSITIVE LOGITS
    lapsingToolbar
    0.61
    Gilla
    0.58
     الرياضيه
    0.56
     للاسماء
    0.55
     uVar
    0.54
     Италијани
    0.53
    FieldBuilder
    0.52
    _));
    0.50
     ब्रेकडाउन
    0.50
    htu
    0.48
    Activations Density 0.379%

    No Known Activations