© 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. 88805
    Prev
    Next
    INDEX
    Explanations

    Code or math snippets

    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
    agic
    -0.29
     mixed
    -0.27
     Mixed
    -0.25
    AGIC
    -0.24
    aria
    -0.24
    æĮ¯åħ´
    -0.24
     communication
    -0.24
    .py
    -0.24
    acob
    -0.24
    etur
    -0.24
    POSITIVE LOGITS
    ä¼ļéķ¿
    0.28
    åIJİæİĴ
    0.26
    è¿ĺä¸įå¤Ł
    0.26
    disposed
    0.26
     Ceiling
    0.26
    .outer
    0.25
    çĨµ
    0.25
    æŃ£åĵģ
    0.25
    ä½ıæīĢ
    0.25
    åĪĪ
    0.25
    Activations Density 0.057%

    No Known Activations