© 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. 64203
    Prev
    Next
    INDEX
    Explanations

    Instructions, How-to's

    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
    å½Ĺ
    -0.28
    æIJĸ
    -0.27
     jig
    -0.25
    éĢ¡
    -0.25
    NBC
    -0.25
    nock
    -0.24
    lcd
    -0.24
     Jacqu
    -0.24
     Lancaster
    -0.24
    _LCD
    -0.24
    POSITIVE LOGITS
    ·»
    0.31
    缴
    0.27
    带æĿ¥çļĦ
    0.26
    åĩºæĿ¥çļĦ
    0.25
    خش
    0.25
    .slim
    0.25
    -conf
    0.25
    privation
    0.25
     fer
    0.25
    åIJĮæł·çļĦ
    0.24
    Activations Density 6.675%

    No Known Activations