© 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. 99248
    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
    Negative Logits
    åĬ³åĬ¨èĢħ
    -0.26
     CLOCK
    -0.26
     trụ
    -0.25
    æĮģæľīçļĦ
    -0.24
     clocks
    -0.24
     advocates
    -0.24
    æ´¾
    -0.24
     refs
    -0.24
    æ´Ĺ
    -0.23
    ä¹ĺçĶ¨è½¦
    -0.23
    POSITIVE LOGITS
    è¿Ľè´§
    0.28
    æĸ¹åIJij
    0.28
    orent
    0.27
    orm
    0.26
    ocom
    0.25
    elm
    0.25
    markt
    0.24
    affe
    0.24
     onAnimation
    0.24
    _Msk
    0.24
    Activations Density 0.014%

    No Known Activations

    This feature has no known activations.