© 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. 31-TRANSCODER-HP
    7. 17911
    Prev
    Next
    INDEX
    Explanations

    pregnancy

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     practiced
    -0.27
    æķı
    -0.26
     bundle
    -0.25
    æĪĴ
    -0.25
    åĽ½ä¼ģ
    -0.24
    çŁ¾
    -0.24
    èIJ½åĪ°å®ŀå¤Ħ
    -0.24
    orious
    -0.23
    .setLayout
    -0.23
     welt
    -0.23
    POSITIVE LOGITS
    åıŁ
    0.28
    swer
    0.28
    iu
    0.27
     conceive
    0.27
    iosa
    0.27
    åħŃ
    0.26
     bear
    0.26
     Six
    0.26
    Six
    0.26
     bears
    0.26
    Activations Density 0.924%

    No Known Activations