© 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. 107838
    Prev
    Next
    INDEX
    Explanations

    undertake

    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
    çĢı覽
    -0.28
    itemid
    -0.27
    ä¸įèī¯ä¿¡æģ¯
    -0.27
    atisfied
    -0.25
     BigInt
    -0.25
    hots
    -0.24
    enames
    -0.24
    SHOT
    -0.24
    idl
    -0.24
    è´Ń票
    -0.24
    POSITIVE LOGITS
     goal
    0.31
    ony
    0.29
    goal
    0.27
    诱
    0.26
    Goal
    0.25
    延éķ¿
    0.25
    éĩį
    0.25
     ÑĢоли
    0.25
    gnu
    0.24
     gar
    0.24
    Activations Density 14.866%

    No Known Activations