© 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. 97108
    Prev
    Next
    INDEX
    Explanations

    loop

    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
    Negative Logits
    ä¸ĢæŶ
    -0.27
    iling
    -0.25
    ^K
    -0.25
    _REPO
    -0.25
    idding
    -0.24
    ëĿ¼ëıĦ
    -0.24
     Bayer
    -0.24
    apur
    -0.23
    ä¸Ĭ游
    -0.23
    ä¸Ģå¹ķ
    -0.23
    POSITIVE LOGITS
    size
    0.28
    istic
    0.27
     conc
    0.25
    cedes
    0.25
    side
    0.25
    chart
    0.25
    lama
    0.25
    istics
    0.24
    shade
    0.24
     performs
    0.24
    Activations Density 0.083%

    No Known Activations