© 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. 63358
    Prev
    Next
    INDEX
    Explanations

    www

    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
    inha
    -0.29
    åıĪæľī
    -0.29
    åŁºç¡Ģ
    -0.25
    åį°è±¡
    -0.24
    неÑĨ
    -0.24
     advance
    -0.24
    åºĶçĶ¨
    -0.24
    æķĻæ¡Ī
    -0.24
     break
    -0.23
    第ä¸Ģ家
    -0.23
    POSITIVE LOGITS
    åĬ¨åijĺ
    0.29
     noe
    0.27
    uckle
    0.26
    eced
    0.25
    åįģåĩł
    0.24
    [__
    0.23
    (sr
    0.23
    ç½ijåıĭè¯Ħ论
    0.23
    æĴij
    0.23
    LU
    0.23
    Activations Density 8.356%

    No Known Activations