© 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. 106894
    Prev
    Next
    INDEX
    Explanations

    the

    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
    Not in Any Lists

    No Comments

    Negative Logits
    æĹłæĦı
    -0.27
     Hers
    -0.26
    OrDefault
    -0.25
    æĻļä¼ļ
    -0.25
    èµ°åħ¥
    -0.25
    ULATOR
    -0.24
    æĻ¯
    -0.24
    eor
    -0.24
    åħĪ导
    -0.24
    æ·¹
    -0.23
    POSITIVE LOGITS
    ORN
    0.27
    ims
    0.26
    çĤ¹äºĨ
    0.25
    plotlib
    0.25
    Cars
    0.25
     familiar
    0.25
    uffy
    0.24
    nio
    0.23
     CONCAT
    0.23
    UILTIN
    0.23
    Activations Density 6.071%

    No Known Activations