© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 18980
    Prev
    Next
    INDEX
    Explanations

    infect, viral, online

    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
    Not in Any Lists

    No Comments

    Negative Logits
    æı¡
    -0.32
    éĺĪ
    -0.29
    å¥Ī
    -0.27
    æ´Ľ
    -0.26
    åħĭæĭī
    -0.26
    æİĪæĿĥ
    -0.26
    åĵij
    -0.26
    aji
    -0.25
    éĶIJ
    -0.25
    éĹ´éļĶ
    -0.25
    POSITIVE LOGITS
     infect
    0.33
     viral
    0.32
     vir
    0.31
    vir
    0.31
    online
    0.31
     virus
    0.30
     online
    0.30
     viruses
    0.29
    ç§ijæĬĢèĤ¡ä»½
    0.29
     circulated
    0.28
    Activations Density 0.748%

    No Known Activations