© 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. 25-TRANSCODER-HP
    7. 106994
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æ³µ
    -0.29
     pump
    -0.27
     "'";↵
    -0.25
    atis
    -0.25
    足
    -0.25
     GUIDE
    -0.24
    ä¸Ģæĸ¹
    -0.24
    ä¸įåIJĮ
    -0.24
    æµ·
    -0.23
    ç®Ģ
    -0.23
    POSITIVE LOGITS
    æī¹éĩı
    0.26
    clid
    0.25
    ucas
    0.25
    äºļ马éĢĬ
    0.25
    “But
    0.24
    OTO
    0.23
    scriptions
    0.23
     ãĥĪãĥ©ãĥĥãĤ¯ãĥIJãĥĥãĤ¯
    0.23
     Incoming
    0.23
     +#+
    0.22
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.