© 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. 19865
    Prev
    Next
    INDEX
    Explanations

    code snippets/technical language

    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
    ész
    -0.27
    æĿ¿
    -0.26
    .Automation
    -0.25
    æºIJ
    -0.25
    filesystem
    -0.24
    anagan
    -0.24
     provid
    -0.24
     strength
    -0.23
    éĤ¦
    -0.23
    ashi
    -0.23
    POSITIVE LOGITS
     hoÃłng
    0.26
    åŀĤ
    0.26
    å¤įåį°
    0.25
    _absolute
    0.25
    èݼ
    0.25
    é¸
    0.25
    /sites
    0.25
     ÄijÆ°Æ¡ng
    0.25
     Absolute
    0.25
    joint
    0.25
    Activations Density 0.106%

    No Known Activations