© 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. 97070
    Prev
    Next
    INDEX
    Explanations

    pregnancy/infant related

    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
    çŁ¥è¯Ĩ
    -0.26
    large
    -0.26
     Downs
    -0.26
    .Pos
    -0.25
    ëŀľ
    -0.25
    æķĻåѦ
    -0.25
    家åºŃæķĻèĤ²
    -0.24
    ropic
    -0.24
    è¿Ļ份
    -0.24
    éĹºèľľ
    -0.23
    POSITIVE LOGITS
    èĶļ
    0.26
    ulet
    0.26
    ält
    0.26
    æĿµ
    0.26
     vero
    0.26
    ARC
    0.25
     adul
    0.25
    èĹ¿
    0.25
    anes
    0.24
    éĴ§
    0.24
    Activations Density 0.106%

    No Known Activations