© 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. 129382
    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
    kea
    -0.27
    ç³į
    -0.27
    urement
    -0.25
    åīľ
    -0.25
     enjo
    -0.24
    RATION
    -0.24
    çļĦåľ°
    -0.23
    èĥĻ
    -0.23
    illez
    -0.23
    ruc
    -0.23
    POSITIVE LOGITS
    do
    0.26
    mind
    0.25
    çľģ
    0.25
    âĢ¦↵
    0.24
    æĢ§åij½
    0.24
    Cop
    0.23
    bo
    0.23
    صرÙģ
    0.23
    åºĦ
    0.23
     rig
    0.23
    Activations Density 0.031%

    No Known Activations

    This feature has no known activations.