© 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. 96993
    Prev
    Next
    INDEX
    Explanations

    Xcode and programming

    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
    ampie
    -0.27
    iton
    -0.27
    andan
    -0.27
    WithPath
    -0.26
    udded
    -0.26
    å®ģæĦ¿
    -0.25
    endir
    -0.25
    Band
    -0.24
     Dot
    -0.24
    éĢļè¡Į
    -0.24
    POSITIVE LOGITS
    COPE
    0.33
    æĥħ
    0.31
    grams
    0.29
    æĹłèģĬ
    0.25
    apping
    0.24
    æ±Ŀ
    0.24
    apped
    0.24
    åİŁèijĹ
    0.24
    ér
    0.24
    losing
    0.24
    Activations Density 1.033%

    No Known Activations