© 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. 53036
    Prev
    Next
    INDEX
    Explanations

    Top position

    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
    rices
    -0.26
    以æѤ
    -0.25
    è¡Įæĥħ
    -0.25
    kish
    -0.25
    Coroutine
    -0.24
    ä½ľèĢħæīĢæľī
    -0.24
    pectrum
    -0.24
     BASIS
    -0.24
    riott
    -0.24
    ichert
    -0.24
    POSITIVE LOGITS
    Top
    0.32
     cong
    0.31
    .top
    0.31
     top
    0.30
     lay
    0.29
    top
    0.28
     Top
    0.28
     VBox
    0.28
     hooked
    0.27
    _top
    0.27
    Activations Density 0.015%

    No Known Activations