© 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. 64000
    Prev
    Next
    INDEX
    Explanations

    programming

    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
    追赶
    -0.27
    åĬłå¿«
    -0.26
    赶
    -0.25
     ë§ŀ
    -0.25
    sembler
    -0.24
    åİ¢
    -0.24
     sociales
    -0.24
    stadt
    -0.24
     göre
    -0.24
     based
    -0.23
    POSITIVE LOGITS
    骨头
    0.26
    bury
    0.26
    DOMNode
    0.25
    ç§ĥ
    0.25
    º«
    0.25
     liá»ģn
    0.24
    Wall
    0.23
    ÑĢеÑģÑĤ
    0.23
    åħ´èĩ´
    0.23
    åģ»
    0.23
    Activations Density 1.697%

    No Known Activations