© 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. 110914
    Prev
    Next
    INDEX
    Explanations

    Start of articles/descriptions

    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
    eness
    -0.30
    unt
    -0.28
    lus
    -0.27
    è¿Ļäºĭ
    -0.27
     ÑĢеÑĪа
    -0.26
    imum
    -0.25
    ius
    -0.25
    let
    -0.25
     fu
    -0.25
    cus
    -0.25
    POSITIVE LOGITS
    Originally
    0.28
    edBy
    0.28
    éķ¿çŁŃ
    0.27
     originally
    0.27
    arov
    0.27
     LinearGradient
    0.27
    opaque
    0.26
    ,sizeof
    0.26
    неÑģен
    0.26
    ardo
    0.26
    Activations Density 0.008%

    No Known Activations