© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 112638
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    æ¯ĭ
    -0.30
    åѤ
    -0.26
     suffix
    -0.26
    ten
    -0.26
    ######
    -0.24
    èĬ±åĽŃ
    -0.24
     conjug
    -0.23
    ->__
    -0.23
     Guinea
    -0.23
    to
    -0.23
    POSITIVE LOGITS
    intval
    0.25
    ä¸įå°ı
    0.24
    apl
    0.24
    etical
    0.24
    æľĢæĸ°
    0.24
    .NewRequest
    0.24
    建
    0.23
    主èIJ¥
    0.23
    uing
    0.23
     reco
    0.23
    Activations Density 0.004%

    No Known Activations

    This feature has no known activations.