© 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. 14-TRANSCODER-HP
    7. 39664
    Prev
    Next
    INDEX
    Explanations

    Abbreviations and codes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_14.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.14.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æĸ°èĤ¡
    -0.28
    OTT
    -0.27
     Eag
    -0.27
    游æĪıçݩ家
    -0.26
    ning
    -0.26
    rength
    -0.25
    å°ıæŶåĢĻ
    -0.25
    ÙĪرة
    -0.25
    mers
    -0.25
    ousse
    -0.24
    POSITIVE LOGITS
     tat
    0.28
    [-
    0.27
    open
    0.27
    vsp
    0.26
    å¼Ģåĩº
    0.26
    å¼ĢæĶ¾
    0.25
    unnable
    0.25
    оÑģÑĤи
    0.24
    èģĶ
    0.24
    å®ĺåı¸
    0.24
    Activations Density 0.275%

    No Known Activations