© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 18-TRANSCODER-HP
    7. 89043
    Prev
    Next
    INDEX
    Explanations

    race and culture

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    icum
    -0.34
    uvw
    -0.29
    imation
    -0.27
     ÑĢа
    -0.25
    semi
    -0.25
    ë²Ķ
    -0.25
    ouv
    -0.25
    Ñĥв
    -0.25
     semi
    -0.25
     Citizen
    -0.24
    POSITIVE LOGITS
    etime
    0.25
    æ³ī
    0.25
    éļħ
    0.25
     aberr
    0.24
    pt
    0.24
    feld
    0.23
    äºŃ
    0.23
    à¸ģะ
    0.23
     colours
    0.23
    è¾¾æłĩ
    0.23
    Activations Density 12.902%

    No Known Activations