© 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. 19722
    Prev
    Next
    INDEX
    Explanations

    Software

    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
    ylon
    -0.35
    èī²
    -0.32
    7
    -0.31
    yx
    -0.29
    é¦Ĩ
    -0.28
    gesch
    -0.27
    ndef
    -0.27
    RuntimeException
    -0.25
     GM
    -0.25
    ç¼ĸ
    -0.24
    POSITIVE LOGITS
     Woody
    0.27
    Aud
    0.27
    customer
    0.26
    hone
    0.26
     overnight
    0.25
    mode
    0.25
    ovel
    0.25
    娥
    0.25
     Advantage
    0.25
    appendTo
    0.24
    Activations Density 0.060%

    No Known Activations