© 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. 142754
    Prev
    Next
    INDEX
    Explanations

    Parenthesis

    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
    nish
    -0.27
    uvwxyz
    -0.27
     Gauss
    -0.26
    åı¤ä»Ĭ
    -0.26
     yourselves
    -0.26
    æĹłæīĢè°ĵ
    -0.25
    eree
    -0.24
     vi
    -0.24
     terminals
    -0.24
    æİĤ
    -0.24
    POSITIVE LOGITS
     xsi
    0.27
    ç½ij讯
    0.27
     zar
    0.26
     Turning
    0.26
    æĿ¥è®²
    0.25
    \core
    0.25
    rats
    0.25
    Turning
    0.24
    rical
    0.24
    arent
    0.24
    Activations Density 25.337%

    No Known Activations