© 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. 88755
    Prev
    Next
    INDEX
    Explanations

    computer code/commands/publications

    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
    igators
    -0.28
    melon
    -0.27
    iram
    -0.26
    /el
    -0.25
    puted
    -0.25
    igator
    -0.24
    è¿Ļç§įæĥħåĨµ
    -0.24
    falls
    -0.24
    tag
    -0.24
    arro
    -0.24
    POSITIVE LOGITS
    带æĿ¥
    0.25
    ä¾Ŀèµĸ
    0.25
    ..."↵↵
    0.25
    å®ī稳
    0.24
     obey
    0.24
    éĤĢ
    0.24
     depended
    0.24
     feder
    0.23
    åħ¥åľº
    0.23
     depends
    0.23
    Activations Density 0.000%

    No Known Activations