© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 130123
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    Not in Any Lists

    No Comments

    Negative Logits
     gamm
    -0.28
    aghetti
    -0.28
    okin
    -0.25
    зов
    -0.24
     æŁ¥
    -0.24
    æĭ¨æīĵ
    -0.24
    æĸ°éĹ»ç½ij
    -0.23
    è¾Ł
    -0.23
    plist
    -0.23
    aden
    -0.23
    POSITIVE LOGITS
     natural
    0.26
    å»īæĶ¿
    0.25
    vore
    0.24
    å»ī
    0.24
    .native
    0.24
    вл
    0.24
    çľ¼çľĭ
    0.24
    è¿ĩ
    0.24
    ä¼łæĿ¥
    0.24
    åį°è±¡
    0.23
    Activations Density 0.004%

    No Known Activations

    This feature has no known activations.