© 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. 30-TRANSCODER-HP
    7. 96693
    Prev
    Next
    INDEX
    Explanations

    code/documentation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ,Object
    -0.26
    çĤ¹å¤´
    -0.25
    assage
    -0.25
    ymax
    -0.24
    ,\"
    -0.24
     outline
    -0.24
    çıłå®Ŀ
    -0.24
    ums
    -0.24
    aign
    -0.24
    à¹Īาย
    -0.24
    POSITIVE LOGITS
    come
    0.32
     Come
    0.29
     come
    0.28
     found
    0.27
    Found
    0.27
    æĦı
    0.26
     Found
    0.26
    ä¸¥æł¼
    0.26
     comes
    0.26
    Come
    0.25
    Activations Density 0.881%

    No Known Activations