© 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. 19848
    Prev
    Next
    INDEX
    Explanations

    long

    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
    íĴį
    -0.28
    æĤ£
    -0.26
    plits
    -0.26
    èIJ½åı¶
    -0.25
    æĽ¾ç»ı
    -0.24
    åģ¥åº·çļĦ
    -0.24
    odore
    -0.23
    .Border
    -0.23
    ç»Ļ她
    -0.23
    utch
    -0.23
    POSITIVE LOGITS
    èģĮå·¥
    0.31
    ities
    0.30
    iaux
    0.26
    罩
    0.26
    cles
    0.25
    ector
    0.25
     Slot
    0.25
    acios
    0.25
    neys
    0.25
    âİŀ
    0.25
    Activations Density 0.985%

    No Known Activations