© 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. 30-TRANSCODER-HP
    7. 95098
    Prev
    Next
    INDEX
    Explanations

    Plaintiff/Appellant

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ADB
    -0.28
    å¾ĹçĽĬäºİ
    -0.27
    ror
    -0.26
    #__
    -0.26
    .tie
    -0.26
    .rgb
    -0.25
    ä¸įåĩºæĿ¥
    -0.24
    balances
    -0.24
     balances
    -0.24
    ยว
    -0.24
    POSITIVE LOGITS
    ocial
    0.30
     german
    0.30
     Hosting
    0.28
    åħ³éŨ
    0.27
    oreal
    0.26
    Çĺ
    0.25
    itag
    0.25
    社åĮº
    0.25
    æĦıè§ģ
    0.24
    leneck
    0.24
    Activations Density 0.100%

    No Known Activations