© 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. 97106
    Prev
    Next
    INDEX
    Explanations

    guide

    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
     brand
    -0.38
    åĵģçīĮ
    -0.37
     studio
    -0.37
     brands
    -0.33
    brand
    -0.33
     Studio
    -0.31
    åĵģçīĮçļĦ
    -0.31
    ãĥĸãĥ©ãĥ³ãĥī
    -0.29
     street
    -0.28
     startup
    -0.28
    POSITIVE LOGITS
     guides
    0.40
     guide
    0.40
     Guide
    0.35
     Guides
    0.35
    endar
    0.34
    guide
    0.32
    -guide
    0.32
    Guide
    0.30
    ãĤ¬ãĤ¤ãĥī
    0.29
    _guide
    0.28
    Activations Density 1.030%

    No Known Activations