© 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. 112880
    Prev
    Next
    INDEX
    Explanations

    Code formatting

    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
     wisely
    -0.30
    /native
    -0.26
    ä¸ļ
    -0.25
     tand
    -0.24
     afterward
    -0.24
     fetched
    -0.23
     seams
    -0.23
    oker
    -0.23
    partials
    -0.23
    doors
    -0.23
    POSITIVE LOGITS
    ä¸Ģä¸ĸ
    0.27
    ’T
    0.26
    åħĦ
    0.26
     Abb
    0.25
    æĢľ
    0.24
    ิà¸ģ
    0.24
    -Level
    0.23
    å¹³éĿ¢
    0.23
     Name
    0.23
    atte
    0.23
    Activations Density 0.005%

    No Known Activations