© 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. 98841
    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
    ä¾
    -0.27
    çļĩåĨł
    -0.26
    æİ¨è¿Ľ
    -0.26
    ละ
    -0.25
    istar
    -0.24
    å¨ģåĬĽ
    -0.24
    è¿Ľå±ķæĥħåĨµ
    -0.24
    istan
    -0.23
    ä¼łè¨Ģ
    -0.23
     Shelby
    -0.23
    POSITIVE LOGITS
    -aff
    0.28
    fty
    0.24
    abble
    0.24
    æľĶ
    0.24
    éĥģ
    0.24
    é¦ĸä½į
    0.23
    MAT
    0.23
    ugar
    0.23
    æ±Ĺ
    0.23
    ugin
    0.23
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.