© 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. 25-TRANSCODER-HP
    7. 108362
    Prev
    Next
    INDEX
    Explanations

    should

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ç´§å¯Ĩç»ĵåIJĪ
    -0.27
    eva
    -0.25
    æĿ¥è®¿
    -0.24
     ana
    -0.24
    æľ¬çĿĢ
    -0.24
    pdb
    -0.24
    presso
    -0.24
    åIJ«
    -0.23
     lagi
    -0.23
    æľīä¸Ģå®ļ
    -0.23
    POSITIVE LOGITS
    ä¸įåºĶ该
    0.30
    æĩī該
    0.29
     shouldn
    0.28
    .should
    0.28
    æĩĤ
    0.28
    should
    0.28
    éħ¬
    0.26
    缨
    0.26
    马ä¸ģ
    0.26
     Should
    0.25
    Activations Density 1.136%

    No Known Activations