© 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. 96985
    Prev
    Next
    INDEX
    Explanations

    Based

    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
    Negative Logits
     on
    -0.30
     auf
    -0.27
    ’on
    -0.27
    çIJĨ论ä¸Ĭ
    -0.26
    çļĦäºĭ
    -0.26
    çī©
    -0.26
    asonry
    -0.26
    åı¦ä¸Ģæĸ¹éĿ¢
    -0.25
    iked
    -0.25
    ROUP
    -0.25
    POSITIVE LOGITS
    runner
    0.29
    LESS
    0.28
    trag
    0.28
    nav
    0.28
    ä¸Ĭ课
    0.27
     prosecute
    0.26
    spl
    0.26
    UpDown
    0.25
    nosis
    0.24
     Spl
    0.24
    Activations Density 0.004%

    No Known Activations