© 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. 95147
    Prev
    Next
    INDEX
    Explanations

    <html

    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
     center
    -0.46
     centre
    -0.44
    ä¸Ńå¿ĥ
    -0.43
    centre
    -0.42
    Centre
    -0.38
    center
    -0.38
    CENTER
    -0.37
    Center
    -0.37
     Center
    -0.36
     centres
    -0.35
    POSITIVE LOGITS
     Cush
    0.29
    ingu
    0.29
    mania
    0.27
    flix
    0.25
    ned
    0.25
    è§Ģå¯Ł
    0.25
    nage
    0.24
    æŃ£æĺ¯
    0.24
    ENCH
    0.24
     ordered
    0.24
    Activations Density 0.103%

    No Known Activations