© 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. 63911
    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
    å¸Ŀçİĭ
    -0.31
    SSIP
    -0.29
     viewpoints
    -0.28
    çĻ½éĽª
    -0.28
    xFE
    -0.27
    è§ĤçĤ¹
    -0.26
     viewing
    -0.26
     events
    -0.25
     teaspoons
    -0.25
    xfe
    -0.25
    POSITIVE LOGITS
    侵害
    0.27
    foundation
    0.27
    (encoder
    0.25
    ahan
    0.25
    使çĶ¨æĿĥ
    0.24
    nic
    0.24
    antu
    0.23
    åı½
    0.23
     jus
    0.23
    جÙħÙĦ
    0.23
    Activations Density 4.160%

    No Known Activations

    This feature has no known activations.