© 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. 107561
    Prev
    Next
    INDEX
    Explanations

    prepositions

    np_max-act · gemini-2.0-flash
    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
     enqu
    -0.26
    å¯Ħæīĺ
    -0.26
     Tribal
    -0.26
    OrderBy
    -0.25
    æĭ¥
    -0.24
    WI
    -0.24
    æĵģ
    -0.24
     circumcision
    -0.23
    .addObject
    -0.23
     UFC
    -0.23
    POSITIVE LOGITS
    ption
    0.29
     retire
    0.27
     Kur
    0.27
    é¾Ħ
    0.26
    .fx
    0.26
    åĴĮæĶ¯æĮģ
    0.26
     shorter
    0.26
    _READONLY
    0.25
    iscard
    0.25
     retired
    0.25
    Activations Density 24.628%

    No Known Activations