© 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. 141809
    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
    åľ¨åºĬä¸Ĭ
    -0.30
    plotlib
    -0.29
    Gre
    -0.28
    çī¯
    -0.27
    èĥ½åľ¨
    -0.27
    odox
    -0.26
    çľĭçĤ¹
    -0.26
    [".
    -0.25
    èĥ½å¤Łåľ¨
    -0.24
    abox
    -0.24
    POSITIVE LOGITS
    çıŃ
    0.29
    åģı
    0.27
    温
    0.27
    æŧ
    0.26
    å·²
    0.26
    ết
    0.25
    ç«Ļ
    0.25
    adic
    0.25
    éĻĦ
    0.25
    级
    0.25
    Activations Density 4.724%

    No Known Activations

    This feature has no known activations.