© 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. 22-TRANSCODER-HP
    7. 129999
    Prev
    Next
    INDEX
    Explanations

    business

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ansa
    -0.27
    æĺ§
    -0.26
    Purpose
    -0.25
    пан
    -0.24
    =""><
    -0.24
    ;/*
    -0.24
     Fully
    -0.24
    ANTITY
    -0.23
     Wig
    -0.23
    usage
    -0.23
    POSITIVE LOGITS
    å¼ĢäºĨ
    0.31
     secret
    0.30
    rone
    0.30
    åĽŀäºĭ
    0.27
    ç§ĺ
    0.27
    l
    0.26
    ário
    0.26
    rze
    0.24
    en
    0.24
    .reduce
    0.24
    Activations Density 0.002%

    No Known Activations