© 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. 108068
    Prev
    Next
    INDEX
    Explanations

    Data, science, math, business

    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
    åİ¿å§Ķ书记
    -0.28
     vocal
    -0.28
    irty
    -0.27
    ëĭ¤ê°Ģ
    -0.26
    /rs
    -0.26
    ould
    -0.25
    fos
    -0.25
    ocrats
    -0.24
    ä½ıäºĨ
    -0.24
    aption
    -0.23
    POSITIVE LOGITS
    appen
    0.26
    è¿ĺä¸į
    0.26
    Per
    0.25
    authorization
    0.24
    ctal
    0.24
     Swe
    0.23
    chie
    0.23
     Cly
    0.23
     Per
    0.23
    NoSuch
    0.23
    Activations Density 28.457%

    No Known Activations