© 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. 33-TRANSCODER-HP
    7. 53513
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oom
    -0.26
    agn
    -0.26
    æĬ½
    -0.26
    :{
    -0.25
    éĤ°
    -0.25
    çļĦå®ŀéĻħ
    -0.24
    人æŃ»äº¡
    -0.24
    ulsion
    -0.24
    }.{
    -0.24
     nomin
    -0.24
    POSITIVE LOGITS
     bur
    0.27
    ERY
    0.26
     Bris
    0.25
    æĢĢæĬ±
    0.25
    ByKey
    0.24
    ä½Ĩæĺ¯å¯¹äºİ
    0.24
    Nam
    0.24
    .central
    0.24
    æĸ°éĹ»ç½ij
    0.23
    çĭIJ
    0.23
    Activations Density 0.005%

    No Known Activations

    This feature has no known activations.