© 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. 25-TRANSCODER-HP
    7. 109355
    Prev
    Next
    INDEX
    Explanations

    regarding

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    示æĦı
    -0.27
    tur
    -0.24
    åĽĽçº§
    -0.24
    Narr
    -0.24
     realise
    -0.23
    _audit
    -0.23
    èģĮ
    -0.23
    ponce
    -0.23
     synopsis
    -0.23
    -half
    -0.23
    POSITIVE LOGITS
    ä¸Ģå®ļçļĦ
    0.27
    avored
    0.27
    ä¸Ģå®ļ
    0.26
     ли
    0.25
     both
    0.25
    ä»ħ次äºİ
    0.25
    ...)↵↵
    0.25
    ...)
    0.25
     &)
    0.25
    ayı
    0.24
    Activations Density 0.306%

    No Known Activations