© 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. 95096
    Prev
    Next
    INDEX
    Explanations

    gibberish

    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
     win
    -0.27
    ç´IJ
    -0.27
     ste
    -0.27
    ä¸įè§ģ
    -0.26
    -cycle
    -0.26
    STRACT
    -0.25
     sh
    -0.25
    ارÙĬ
    -0.25
     wood
    -0.25
     asc
    -0.24
    POSITIVE LOGITS
    capture
    0.27
    产åѦ
    0.27
    Ahead
    0.26
    Already
    0.26
    ä¸īåĪĨ
    0.26
    _cross
    0.26
     bookmarks
    0.25
    cepts
    0.25
    captures
    0.25
     cpt
    0.25
    Activations Density 0.882%

    No Known Activations