© 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. 33-TRANSCODER-HP
    7. 53508
    Prev
    Next
    INDEX
    Explanations

    Recently

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸įç®¡ä½ł
    -0.29
    ä¸įèĥ½ä¸ºç©º
    -0.27
    ascript
    -0.27
    为æĬĵæīĭ
    -0.26
    :''
    -0.26
    alo
    -0.26
    æĺ¯åı¯ä»¥
    -0.25
    èĩ³æĸ¼
    -0.25
    ç»Ļä»ĸ们
    -0.25
     doctrines
    -0.25
    POSITIVE LOGITS
     first
    0.42
     available
    0.36
     arranged
    0.35
     First
    0.35
    ,
    0.34
     Available
    0.32
    first
    0.31
     added
    0.31
     scheduled
    0.31
     outlined
    0.31
    Activations Density 0.033%

    No Known Activations