© 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. 95043
    Prev
    Next
    INDEX
    Explanations

    Code or instructions

    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
    æĪ¾
    -0.30
    -period
    -0.28
    åĨ»
    -0.27
    çĥĬ
    -0.26
    èĬĿ麻
    -0.26
    沸
    -0.26
    èħ¥
    -0.25
    湯
    -0.25
    æĪĺ
    -0.25
     matte
    -0.25
    POSITIVE LOGITS
    rops
    0.25
     Manus
    0.25
     corner
    0.25
    èµ°å»Ĭ
    0.25
    åĨĬ
    0.24
     realistically
    0.24
    ULO
    0.24
    picked
    0.24
    ched
    0.24
    hall
    0.24
    Activations Density 0.003%

    No Known Activations