© 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. 88883
    Prev
    Next
    INDEX
    Explanations

    Adjectives describing fields

    np_max-act · gemini-2.0-flash
    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
     tail
    -0.28
    家éŨåı£
    -0.24
     iceberg
    -0.24
    vil
    -0.24
     directly
    -0.23
     household
    -0.23
     head
    -0.23
     extr
    -0.23
     Disposable
    -0.23
    _tail
    -0.23
    POSITIVE LOGITS
     swung
    0.29
     causa
    0.28
     opendir
    0.27
     cach
    0.27
    ä¹ħ
    0.27
    alah
    0.25
    ä¸ĢéĿ¢
    0.23
    åĩºä¼ļ
    0.23
    潦
    0.23
    Ĥ¬
    0.23
    Activations Density 0.012%

    No Known Activations