© 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. 31-TRANSCODER-HP
    7. 17798
    Prev
    Next
    INDEX
    Explanations

    code or technical language

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Nad
    -0.27
     nag
    -0.27
     Nag
    -0.25
     race
    -0.25
    letics
    -0.25
    ÑĤÑĢаÑĤ
    -0.24
    迪æĭľ
    -0.24
    alt
    -0.23
    usive
    -0.23
     век
    -0.23
    POSITIVE LOGITS
    duc
    0.29
    综åIJĪæķ´æ²»
    0.29
     Sweep
    0.27
    duce
    0.27
    ĮĢ
    0.27
     cy
    0.26
    æĿľ
    0.25
     sweep
    0.25
     lu
    0.25
    lix
    0.24
    Activations Density 1.066%

    No Known Activations