© 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. 89687
    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
    ropic
    -0.24
    責
    -0.24
     estim
    -0.24
    icit
    -0.24
     mint
    -0.23
     station
    -0.23
    ,ep
    -0.23
    西åįĹ
    -0.23
     dispatch
    -0.23
     its
    -0.23
    POSITIVE LOGITS
    cape
    0.28
    ç»Ļä½łä»¬
    0.28
    æ£į
    0.27
     Booth
    0.26
    LinkId
    0.26
     chrono
    0.26
     boilers
    0.25
    shelf
    0.25
     amp
    0.24
    amp
    0.24
    Activations Density 0.831%

    No Known Activations

    This feature has no known activations.