© 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. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 108103
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yses
    -0.31
     pl
    -0.28
    OI
    -0.27
    mium
    -0.26
    åĪ«
    -0.26
     hill
    -0.26
    ucer
    -0.26
    ä¸Ģæľ¬ä¹¦
    -0.25
     presentations
    -0.25
     qu
    -0.25
    POSITIVE LOGITS
    å¥Ķ
    0.28
    utsche
    0.26
    举
    0.24
    ä¸įéĢļ
    0.24
    éŁ
    0.24
    bes
    0.24
     gonna
    0.24
     SUCH
    0.23
    (Syntax
    0.23
    æĭħå½ĵ
    0.23
    Activations Density 6.481%

    No Known Activations

    This feature has no known activations.