© 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. 94903
    Prev
    Next
    INDEX
    Explanations

    Port

    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
    æĭĦ
    -0.29
     brace
    -0.28
    -shot
    -0.25
    ils
    -0.25
    stack
    -0.25
     pack
    -0.25
     stack
    -0.24
     case
    -0.24
    (il
    -0.23
    ile
    -0.23
    POSITIVE LOGITS
    éĺ³
    0.28
    epy
    0.27
    çļĦæķ°éĩı
    0.27
    éºĴéºŁ
    0.27
     Royale
    0.26
    assadors
    0.26
    åħ¥
    0.26
    åģķ
    0.26
    implified
    0.25
    eing
    0.25
    Activations Density 0.885%

    No Known Activations