© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 142552
    Prev
    Next
    INDEX
    Explanations

    statistics and quantities

    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
    Negative Logits
    оде
    -0.25
    uds
    -0.25
     Similar
    -0.24
    ;m
    -0.24
    ress
    -0.24
    èĩªåĬ¨åĮĸ
    -0.24
    éĵ¶è¡Į
    -0.24
    ats
    -0.24
    èģĶ
    -0.24
    antan
    -0.24
    POSITIVE LOGITS
    æĥĭ
    0.31
     ultimate
    0.30
    ä¸į幸
    0.29
    è¿Ļä¹Īå¤ļå¹´
    0.28
    ç
    0.26
    ç«ĭå¿Ĺ
    0.26
    iteur
    0.25
     unfortunately
    0.25
    function
    0.25
    æĤ²åī§
    0.25
    Activations Density 2.117%

    No Known Activations