© 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. 31-TRANSCODER-HP
    7. 18443
    Prev
    Next
    INDEX
    Explanations

    long

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    以人æ°ij为
    -0.27
    itious
    -0.27
    RITE
    -0.24
    æĸ°å±ĢéĿ¢
    -0.24
    arkin
    -0.24
    \Post
    -0.24
    å¾Ī大ç¨ĭ度
    -0.24
    沿
    -0.24
    eton
    -0.24
    iset
    -0.24
    POSITIVE LOGITS
    æŶéĹ´
    0.33
     time
    0.33
     æŶéĹ´
    0.32
    ä¹ĭåIJİ
    0.30
     tijd
    0.29
    ä¸įè¿ĩ
    0.29
     into
    0.28
    ahn
    0.28
    anh
    0.28
    æĻĤéĸĵ
    0.28
    Activations Density 0.816%

    No Known Activations