© 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. 20-TRANSCODER-HP
    7. 129437
    Prev
    Next
    INDEX
    Explanations

    programming, code, technical

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    landers
    -0.26
    æľĪä¸ĭ
    -0.26
    agner
    -0.25
    é¢ĨåĨĽ
    -0.24
    captures
    -0.24
     affairs
    -0.24
    Frameworks
    -0.24
    çıŃ
    -0.24
    åıĹ欢è¿İ
    -0.23
    _GRID
    -0.23
    POSITIVE LOGITS
    appe
    0.29
     backdrop
    0.29
     shade
    0.27
     näm
    0.26
    adients
    0.26
    inus
    0.25
     numeral
    0.25
     Rec
    0.24
    ycop
    0.24
    CKET
    0.24
    Activations Density 50.539%

    No Known Activations