© 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. 94915
    Prev
    Next
    INDEX
    Explanations

    Sports

    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
    ohl
    -0.28
    åıĤåĬłä¼ļè®®
    -0.26
     Kron
    -0.26
    idia
    -0.26
    (lambda
    -0.26
    uran
    -0.26
     DOMAIN
    -0.26
    ÑıÑī
    -0.25
    rial
    -0.25
     maj
    -0.25
    POSITIVE LOGITS
    ä½ĵ
    0.33
     setters
    0.30
     national
    0.29
    (ns
    0.28
     delegate
    0.28
    å¥Ń
    0.27
     ns
    0.27
     corner
    0.26
    ate
    0.26
    åĽ½å®¶
    0.26
    Activations Density 1.039%

    No Known Activations