© 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. 43824
    Prev
    Next
    INDEX
    Explanations

    cirrhosis

    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
     Trot
    -0.27
    invoke
    -0.26
     Grove
    -0.25
    cca
    -0.25
    lı
    -0.25
     Rosenstein
    -0.24
    éĢ»è¾ij
    -0.24
     JAXBElement
    -0.23
    rage
    -0.23
    Become
    -0.23
    POSITIVE LOGITS
    =read
    0.27
     vel
    0.27
     hum
    0.25
    奢ä¾Ī
    0.24
     det
    0.24
    udas
    0.24
     Tape
    0.24
    ç¦Ģ
    0.24
    orrh
    0.24
    ãĥģãĥ¥
    0.23
    Activations Density 0.001%

    No Known Activations