© 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. 33-TRANSCODER-HP
    7. 52750
    Prev
    Next
    INDEX
    Explanations

    forced

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    pto
    -0.30
    ç½ijç»ľä¼łæĴŃ
    -0.28
    artner
    -0.27
    FUN
    -0.26
    ủng
    -0.26
    åī§
    -0.25
     grid
    -0.25
     res
    -0.24
     Mang
    -0.24
    è¾¾ä¸įåĪ°
    -0.24
    POSITIVE LOGITS
    ä¼ij
    0.28
     bearings
    0.28
    ector
    0.26
    è½®
    0.25
     Gon
    0.24
     bear
    0.24
     creep
    0.24
    dio
    0.24
     bears
    0.24
    ectors
    0.24
    Activations Density 0.020%

    No Known Activations