© 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. 107626
    Prev
    Next
    INDEX
    Explanations

    surgeries

    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
    管çIJĨåijĺ
    -0.29
    affen
    -0.27
    ç»°
    -0.26
     Editors
    -0.26
     planners
    -0.26
    贯穿
    -0.26
    Credentials
    -0.25
    slash
    -0.25
    ä¸İæŶ
    -0.25
     CONTRIBUT
    -0.25
    POSITIVE LOGITS
     being
    0.29
    (-(
    0.27
     receiver
    0.26
     {-
    0.26
     Reverse
    0.25
    modifiable
    0.25
     reverse
    0.24
    æĽ´èĥ½
    0.24
    able
    0.24
     lens
    0.23
    Activations Density 10.060%

    No Known Activations