© 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. 20-TRANSCODER-HP
    7. 140601
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     im
    -0.27
     independently
    -0.26
    åĢŃ
    -0.26
    ä¼ģ
    -0.26
     oper
    -0.25
    æĪijä»¬åľ¨
    -0.25
    åİ¢
    -0.24
     enslaved
    -0.24
    çĽ¼
    -0.24
    ap
    -0.24
    POSITIVE LOGITS
    éĢĢè´§
    0.29
    oralType
    0.28
     recall
    0.27
    ARIO
    0.27
    court
    0.25
    irts
    0.25
    erra
    0.25
    ForResult
    0.24
    phans
    0.24
    çļĦåľ°
    0.24
    Activations Density 12.779%

    No Known Activations