© 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. 64181
    Prev
    Next
    INDEX
    Explanations

    Technical/Procedural Language

    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
    ramework
    -0.28
    éĢĤ度
    -0.27
     [('
    -0.26
     ann
    -0.26
    ä¸Ģæĸ¹
    -0.26
    dar
    -0.26
    vinc
    -0.25
     farklı
    -0.24
    æĶ¾æīĭ
    -0.24
     modular
    -0.24
    POSITIVE LOGITS
     joined
    0.28
    é¦Ļåij³
    0.25
    Signal
    0.25
    å¿ĹæĦ¿
    0.24
    .cx
    0.24
    åĴĮ社ä¼ļ
    0.24
    âĨIJ
    0.24
    æ±Ģ
    0.24
     <<-
    0.24
    asjon
    0.24
    Activations Density 22.682%

    No Known Activations