© 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. 17822
    Prev
    Next
    INDEX
    Explanations

    cooking

    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
    onda
    -0.27
    olle
    -0.27
    ubi
    -0.26
    enha
    -0.26
    usalem
    -0.26
    ymi
    -0.26
    HB
    -0.25
    ymb
    -0.24
    scan
    -0.24
    对æłĩ
    -0.24
    POSITIVE LOGITS
     cooking
    0.60
     cook
    0.55
    Cook
    0.54
    çĥ¹é¥ª
    0.54
     Cook
    0.54
     recipes
    0.52
     cooks
    0.51
     Cooking
    0.49
    åݨ
    0.49
     cooked
    0.48
    Activations Density 0.011%

    No Known Activations