© 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. 106965
    Prev
    Next
    INDEX
    Explanations

    Visual perception or searching

    np_max-act-logits · gemini-2.0-flash
    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
    æ¹¾
    -0.31
    æĴŀåĩ»
    -0.27
    è¨Ģè¯Ń
    -0.26
     Coleman
    -0.26
    ugin
    -0.25
     Investig
    -0.25
    责任æĦŁ
    -0.25
     i
    -0.25
     touchdowns
    -0.24
    æŀī
    -0.24
    POSITIVE LOGITS
    åħ¨å¹´
    0.30
    Èĺ
    0.26
    为æĤ¨æıIJä¾Ľ
    0.26
    ä¸ĢæĹı
    0.25
    æģŃ
    0.25
    æľ¬ç«Ļ
    0.25
    IJèĹı
    0.25
     guerr
    0.25
    acente
    0.24
    steen
    0.24
    Activations Density 1.043%

    No Known Activations