© 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. 20-TRANSCODER-HP
    7. 131069
    Prev
    Next
    INDEX
    Explanations

    Inventions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    roph
    -0.28
    izer
    -0.26
    ulle
    -0.25
    å¿¡
    -0.24
    rips
    -0.24
    Tro
    -0.24
     наÑģел
    -0.24
    ä¸įå¼Ģ
    -0.24
    ibi
    -0.24
    ÑĪа
    -0.24
    POSITIVE LOGITS
    èĩ»
    0.29
     bid
    0.26
    OLEAN
    0.24
    _alive
    0.24
    建设
    0.24
    —they
    0.24
     cockpit
    0.24
    Smarty
    0.24
     hor
    0.23
    ETCH
    0.23
    Activations Density 4.930%

    No Known Activations