© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 41081
    Prev
    Next
    INDEX
    Explanations

    wouldn

    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
    Negative Logits
    tent
    -0.32
    Release
    -0.28
    mas
    -0.27
     Release
    -0.26
     tent
    -0.26
    elen
    -0.26
    release
    -0.25
     dém
    -0.25
     Tent
    -0.25
    draft
    -0.24
    POSITIVE LOGITS
     conduct
    0.25
     sâu
    0.25
    (rec
    0.25
    ൻ
    0.25
    slice
    0.25
     Hurricanes
    0.24
    积水
    0.24
    promise
    0.23
    éĿ¢çĻ½
    0.23
    æ·±
    0.23
    Activations Density 0.001%

    No Known Activations