© 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. 141212
    Prev
    Next
    INDEX
    Explanations

    Mixed languages and code

    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
    ards
    -0.30
    isex
    -0.29
    én
    -0.28
     жÑĥÑĢ
    -0.26
    orsch
    -0.26
    _lit
    -0.25
    rias
    -0.25
    æĹ¥æĬ¥
    -0.25
     SG
    -0.25
    æĹ¥çĶµ
    -0.25
    POSITIVE LOGITS
    âĨIJ
    0.29
    ä¼ı
    0.26
     vis
    0.26
    gle
    0.26
    #
    0.26
    GLOBALS
    0.25
    ueblo
    0.25
    vis
    0.25
    fs
    0.25
    conf
    0.25
    Activations Density 19.210%

    No Known Activations