© 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. 130206
    Prev
    Next
    INDEX
    Explanations

    aid

    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
    涨价
    -0.28
     Classics
    -0.26
    elight
    -0.26
    æľī人
    -0.26
    tres
    -0.25
    eworld
    -0.25
     Worlds
    -0.25
    ovi
    -0.24
    &S
    -0.24
    éģĵ
    -0.24
    POSITIVE LOGITS
    gnu
    0.27
    强奸
    0.26
    imating
    0.26
    åİŁåĪĻ
    0.25
    eam
    0.25
    /in
    0.24
    vert
    0.24
    eea
    0.24
    马æĭīæĿ¾
    0.24
    èŃ
    0.24
    Activations Density 19.494%

    No Known Activations