© 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. 53062
    Prev
    Next
    INDEX
    Explanations

    code or violent actions

    np_max-act-logits · gemini-2.0-flash
    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
    .tencent
    -0.32
    isible
    -0.31
    bsite
    -0.26
    èĥĨ
    -0.26
    è¯ĹæŃĮ
    -0.25
    mem
    -0.25
    Fan
    -0.25
    fan
    -0.25
    èĨ½
    -0.24
    mist
    -0.24
    POSITIVE LOGITS
    QualifiedName
    0.27
    riage
    0.27
    夯å®ŀ
    0.25
    (Stream
    0.25
     subsidi
    0.25
     rawData
    0.24
     overloaded
    0.24
     Restr
    0.24
    é¦ĸåıij
    0.24
     Sty
    0.24
    Activations Density 0.011%

    No Known Activations