© 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. 30-TRANSCODER-HP
    7. 95813
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æŀĦ建
    -0.37
     build
    -0.33
     constructs
    -0.32
    Build
    -0.30
    build
    -0.30
     Build
    -0.29
    _build
    -0.29
    -build
    -0.28
    建
    -0.28
    PRI
    -0.28
    POSITIVE LOGITS
     table
    0.52
     Table
    0.45
    表
    0.44
     TABLE
    0.40
    table
    0.40
    <table
    0.39
    Table
    0.38
    .table
    0.38
    -table
    0.38
    _table
    0.38
    Activations Density 0.886%

    No Known Activations