© 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. 111209
    Prev
    Next
    INDEX
    Explanations

    digits

    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
    Not in Any Lists

    No Comments

    Negative Logits
    xing
    -0.28
     buckle
    -0.25
     pill
    -0.24
    -digit
    -0.24
     bgcolor
    -0.24
    iry
    -0.24
     waist
    -0.23
    yw
    -0.23
    æĽ´è¦ģ
    -0.23
     brushes
    -0.23
    POSITIVE LOGITS
     Crush
    0.27
    oret
    0.27
    croft
    0.27
    æľºçĶµ
    0.25
    ä¸ĭéĿ¢æĺ¯
    0.25
     дог
    0.25
    -master
    0.24
    odore
    0.24
    åºķä¸ĭ
    0.24
    åįģä½Ļå¹´
    0.24
    Activations Density 0.007%

    No Known Activations