© 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. 141448
    Prev
    Next
    INDEX
    Explanations

    code snippets

    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
    ullo
    -0.28
    éĽĨæĪIJ
    -0.26
     chac
    -0.26
    åľ¨åIJĮä¸Ģ
    -0.26
    妥
    -0.26
     contributed
    -0.26
    该项
    -0.25
    odem
    -0.25
    åħ±åIJĮ
    -0.25
    pora
    -0.24
    POSITIVE LOGITS
    enor
    0.27
    oppins
    0.27
    æµĥ
    0.26
    ransition
    0.26
    >{$
    0.25
    ä¸İåıijå±ķ
    0.24
    /theme
    0.24
     Va
    0.24
    InstanceState
    0.24
    _terminal
    0.24
    Activations Density 74.250%

    No Known Activations