© 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. 32-TRANSCODER-HP
    7. 43814
    Prev
    Next
    INDEX
    Explanations

    medical procedures

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    urbed
    -0.26
     lạ
    -0.26
     counterfeit
    -0.25
    chè
    -0.25
     curt
    -0.24
     '/'↵
    -0.24
    è¿Ľè´§
    -0.24
    URA
    -0.24
     "/"↵
    -0.24
    èĬ±
    -0.24
    POSITIVE LOGITS
     carried
    0.37
     performed
    0.36
     carries
    0.31
     done
    0.30
    çŁ³èĨı
    0.29
     undertaken
    0.29
     reversal
    0.29
     procedures
    0.29
     esl
    0.29
    carry
    0.28
    Activations Density 0.048%

    No Known Activations