© 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. 112509
    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
    rack
    -0.24
    央
    -0.24
    iator
    -0.23
    -alone
    -0.23
    strup
    -0.23
    æ°ijå¿ĥ
    -0.23
    _DECLARE
    -0.23
     Joi
    -0.23
    æĤ¨å¥½
    -0.22
     choke
    -0.22
    POSITIVE LOGITS
    çļĦåı¯èĥ½æĢ§
    0.25
    éĤ¸
    0.24
     Mim
    0.24
    èĢ»
    0.24
    çļĦåħ³æ³¨
    0.24
    anner
    0.23
    preter
    0.23
    åİĨæĿ¥
    0.23
    /tree
    0.22
     Bray
    0.22
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.