© 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. 32-TRANSCODER-HP
    7. 44221
    Prev
    Next
    INDEX
    Explanations

    Broad, varied subjects

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     loop
    -0.29
    Observers
    -0.28
    ä½ľä¸ļ
    -0.27
     Bij
    -0.27
    loop
    -0.26
     Coroutine
    -0.25
    lest
    -0.25
    lds
    -0.24
    vester
    -0.24
    溯
    -0.24
    POSITIVE LOGITS
    æ´ĭ
    0.31
    éĥ¨ä¸»ä»»
    0.28
    æľĿ
    0.27
    éĥ¨åī¯éĥ¨éķ¿
    0.26
     centers
    0.26
    BUILD
    0.25
    ÑĢож
    0.25
     centres
    0.25
     Ñĩелов
    0.25
    ưỡng
    0.25
    Activations Density 0.000%

    No Known Activations