© 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. 19841
    Prev
    Next
    INDEX
    Explanations

    diets, nutrition, deficiencies

    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
    kü
    -0.28
     Presidency
    -0.28
    aida
    -0.28
    éĮ
    -0.27
    inka
    -0.27
    æł¸
    -0.27
     hurt
    -0.26
     clock
    -0.25
    ahn
    -0.25
    iku
    -0.25
    POSITIVE LOGITS
     deprivation
    0.30
    gap
    0.29
    æIJı
    0.29
    -inst
    0.28
    -de
    0.28
    åIJijä¸ĭ
    0.26
     gap
    0.26
    unning
    0.26
    -po
    0.26
    rant
    0.25
    Activations Density 0.224%

    No Known Activations