© 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. 22-TRANSCODER-HP
    7. 111083
    Prev
    Next
    INDEX
    Explanations

    scientific abstracts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Garrett
    -0.26
    æ¶
    -0.26
    reads
    -0.25
     Finished
    -0.25
    anos
    -0.25
     aba
    -0.24
     dissert
    -0.24
    -interest
    -0.24
     oven
    -0.23
    .Constant
    -0.23
    POSITIVE LOGITS
    بت
    0.25
    ä½ŀ
    0.24
    为é¦ĸçļĦ
    0.24
    obot
    0.23
    ä¸Ģä¸ĸ
    0.23
    æĬ±ä½ı
    0.23
     Elect
    0.23
    è¿ĻäºĽäºº
    0.23
    æĿĨèıĮ
    0.23
    éŃij
    0.23
    Activations Density 0.009%

    No Known Activations