© 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. 19-TRANSCODER-HP
    7. 9552
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    é¹Ĭ
    -0.31
     kans
    -0.27
    ä½İä¸ĭ
    -0.27
    arte
    -0.26
    ä¸İä¼Ĺ
    -0.25
    birds
    -0.24
     enorm
    -0.24
    大äºĭ
    -0.24
    ali
    -0.23
     &_
    -0.23
    POSITIVE LOGITS
    ;")↵
    0.28
     Misc
    0.28
    )!↵
    0.25
    ...)↵
    0.25
    ...")↵
    0.25
     Leg
    0.25
    æĥ
    0.24
    !!)↵
    0.24
    imedia
    0.24
    ...";↵
    0.24
    Activations Density 0.080%

    No Known Activations