© 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. 19765
    Prev
    Next
    INDEX
    Explanations

    technical website content

    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
    abies
    -0.28
    bsp
    -0.27
     Babylon
    -0.26
     ters
    -0.26
    AIL
    -0.26
     sÃłng
    -0.25
    ç°Ł
    -0.25
    ottie
    -0.25
    vd
    -0.24
    è¯Ŀ
    -0.24
    POSITIVE LOGITS
    顶
    0.28
    ç§ij
    0.28
    up
    0.28
     procur
    0.28
     Excell
    0.26
     practise
    0.26
    ä¸Ģæŀļ
    0.26
     Bur
    0.25
    uur
    0.25
     practice
    0.25
    Activations Density 0.000%

    No Known Activations