© 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. 19934
    Prev
    Next
    INDEX
    Explanations

    cup

    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
     MASTER
    -0.29
    sheets
    -0.27
    çīº
    -0.27
    sheet
    -0.26
    Past
    -0.26
    bett
    -0.26
     GOODMAN
    -0.26
    ẽ
    -0.25
    sitemap
    -0.25
    è¿Ľé¡¹
    -0.25
    POSITIVE LOGITS
    ios
    0.29
    &w
    0.29
    lag
    0.28
    adol
    0.28
     holder
    0.28
    uka
    0.28
    -holder
    0.27
    lug
    0.27
    asion
    0.27
     hold
    0.26
    Activations Density 0.124%

    No Known Activations