© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 95252
    Prev
    Next
    INDEX
    Explanations

    medical and surgical procedures

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aq
    -0.26
    qv
    -0.25
    ControlItem
    -0.24
    _fh
    -0.24
    绣ä¸Ģ
    -0.24
    llu
    -0.23
    anger
    -0.23
    WA
    -0.23
     evolve
    -0.23
     tamp
    -0.23
    POSITIVE LOGITS
     fib
    0.31
     pain
    0.28
    corn
    0.27
    fib
    0.27
     Hog
    0.26
    æĹłç¼ĺ
    0.26
    skin
    0.26
    æĭĴç»Ŀ
    0.25
    鸡
    0.25
     pul
    0.25
    Activations Density 0.107%

    No Known Activations