© 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. 22-TRANSCODER-HP
    7. 86903
    Prev
    Next
    INDEX
    Explanations

    Instances

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     cryst
    -0.29
    cry
    -0.28
    chè
    -0.27
     borrow
    -0.26
    riers
    -0.26
    ä½Ĩæĺ¯å¯¹äºİ
    -0.26
    äm
    -0.25
    飾
    -0.25
    aniem
    -0.25
    ãĥ³ãĤ°
    -0.25
    POSITIVE LOGITS
    n
    0.28
     here
    0.28
     µ
    0.27
    here
    0.26
    ðŁ¤¨
    0.25
    div
    0.25
     brief
    0.25
     pitched
    0.24
    æľ¬ç«Ļ
    0.24
    åı£
    0.24
    Activations Density 0.001%

    No Known Activations