© 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. 31-TRANSCODER-HP
    7. 18935
    Prev
    Next
    INDEX
    Explanations

    Video games and war

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ve
    -0.30
     field
    -0.27
     bú
    -0.27
     ve
    -0.26
    plug
    -0.25
    curve
    -0.25
    sch
    -0.24
    .Field
    -0.24
    resher
    -0.24
    è¿ĩç¨ĭ
    -0.24
    POSITIVE LOGITS
     GDK
    0.29
     observes
    0.28
     Monitor
    0.28
     Har
    0.26
    ilyn
    0.26
     Gotham
    0.25
    stay
    0.25
    alam
    0.25
    软
    0.25
    é¢Ŀ
    0.25
    Activations Density 0.404%

    No Known Activations