© 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. 18-TRANSCODER-HP
    7. 88592
    Prev
    Next
    INDEX
    Explanations

    Code, programs, or links

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ly
    -0.27
    Core
    -0.27
    ä¸Ģå®ļçļĦ
    -0.27
    gram
    -0.26
    imal
    -0.26
    ker
    -0.26
    missions
    -0.25
     environmental
    -0.25
    yl
    -0.24
     neutral
    -0.24
    POSITIVE LOGITS
    ä¼¼çļĦ
    0.27
    çļĦéĤ£ç§į
    0.27
    çªĺ
    0.26
    姹
    0.25
    .ERR
    0.25
    æ»ļåĬ¨
    0.24
    å©Ģ
    0.24
    >((
    0.24
    åľ¨èģĮ
    0.24
    $action
    0.24
    Activations Density 3.442%

    No Known Activations