© 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. 97109
    Prev
    Next
    INDEX
    Explanations

    Numerical expressions

    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
     Lew
    -0.25
    uras
    -0.25
     targetType
    -0.25
    Gre
    -0.25
     enim
    -0.25
    TÃľ
    -0.24
    anio
    -0.24
    ä¸Ģåı·
    -0.24
    anguage
    -0.24
     draws
    -0.24
    POSITIVE LOGITS
     sop
    0.26
     microtime
    0.26
    trand
    0.26
    æ¹Ħ
    0.26
    åħ¨éĿ¢èIJ½å®ŀ
    0.25
    åĿĽ
    0.24
    -bal
    0.24
    бал
    0.24
     onLoad
    0.24
    izontal
    0.24
    Activations Density 0.884%

    No Known Activations