© 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. 130346
    Prev
    Next
    INDEX
    Explanations

    million

    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
    èIJ½
    -0.37
    èIJ½ä¸ĭ
    -0.30
    äºİ
    -0.28
    ä¸Ģç§Ĵ
    -0.28
    getY
    -0.27
     lapse
    -0.26
    ä¹Łåı¯
    -0.26
    åIJİèĢħ
    -0.25
    åĮĢ
    -0.25
    èĩªç§ģ
    -0.23
    POSITIVE LOGITS
     complication
    0.30
     comp
    0.28
    æ¯Ķå¦Ĥ说
    0.26
    èĥ´
    0.26
    pected
    0.25
     companion
    0.25
     Kont
    0.24
    eki
    0.24
    ç»ŀ
    0.24
     fran
    0.24
    Activations Density 10.615%

    No Known Activations