© 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. 97023
    Prev
    Next
    INDEX
    Explanations

    Google, .com

    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
    山路
    -0.29
    inyin
    -0.29
     Assembly
    -0.28
    plevel
    -0.28
    æĭ¬
    -0.27
    imiter
    -0.27
    éĵ
    -0.26
     commuting
    -0.26
    yan
    -0.26
     assembly
    -0.26
    POSITIVE LOGITS
    ores
    0.32
    esch
    0.30
    UD
    0.28
    Ended
    0.28
    ÅŁÄ±
    0.27
    UDO
    0.26
    è¿ĩæĿ¥
    0.25
    üns
    0.25
    UNG
    0.24
     Ended
    0.24
    Activations Density 0.112%

    No Known Activations