© 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. 88851
    Prev
    Next
    INDEX
    Explanations

    already

    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
    .baomidou
    -0.29
    itional
    -0.27
    åĨħ容ç®Ģä»ĭ
    -0.26
     tÆ°á»Ľng
    -0.25
    æľªæĪIJ
    -0.25
    åĬŀæ³ķ
    -0.24
    ESPN
    -0.24
    still
    -0.23
    Originally
    -0.23
    é¢Ħæ¡Ī
    -0.23
    POSITIVE LOGITS
    дÑĢ
    0.28
    æ²Ĩ
    0.26
    verages
    0.26
    REA
    0.25
     hurt
    0.24
     deeply
    0.24
    å¸Ĥåľºç»ıæµİ
    0.23
     damn
    0.23
    rim
    0.23
     quite
    0.23
    Activations Density 0.101%

    No Known Activations