© 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. 18-TRANSCODER-HP
    7. 88654
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fort
    -0.28
    é£İæļ´
    -0.27
    .LoggerFactory
    -0.27
    lashes
    -0.27
     Fort
    -0.25
     dispos
    -0.25
     Some
    -0.25
    ä¸ĭéĿ¢æĺ¯
    -0.25
     Kron
    -0.24
    æııåĨĻ
    -0.24
    POSITIVE LOGITS
    rat
    0.26
    -pe
    0.26
    缸åħ³æĸ°éĹ»
    0.25
    creens
    0.25
    ilm
    0.25
    å¾Ģå¾Ģä¼ļ
    0.24
    esser
    0.24
    posit
    0.24
    ç»ıèIJ¥ç®¡çIJĨ
    0.24
    iq
    0.23
    Activations Density 2.225%

    No Known Activations

    This feature has no known activations.