© 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. 25-TRANSCODER-HP
    7. 108620
    Prev
    Next
    INDEX
    Explanations

    quotation marks/apostrophes

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     IDE
    -0.33
     Method
    -0.32
     method
    -0.29
     cur
    -0.29
     Development
    -0.27
    è°ı
    -0.27
     methods
    -0.26
    "
    -0.26
    说æĺİ
    -0.26
    whereIn
    -0.26
    POSITIVE LOGITS
    ä¸ĢåĪĢ
    0.28
    arus
    0.27
    ulus
    0.27
    çIJī
    0.27
    åĨįæĿ¥
    0.25
    ático
    0.25
    CKET
    0.25
    缴è¾ĸ
    0.25
     shootout
    0.24
     Pistol
    0.24
    Activations Density 0.908%

    No Known Activations