© 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. 32-TRANSCODER-HP
    7. 40939
    Prev
    Next
    INDEX
    Explanations

    medical patients

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     cor
    -0.29
     becoming
    -0.28
     beaten
    -0.27
    bec
    -0.27
    æĪIJ为
    -0.27
    }->
    -0.26
    æµ®
    -0.25
     become
    -0.25
    çĽ´è¾¾
    -0.25
    adel
    -0.24
    POSITIVE LOGITS
    æīĢéľĢçļĦ
    0.27
    proxy
    0.26
    è¯ģ件
    0.26
    anyl
    0.26
    -browser
    0.25
    elon
    0.25
    eters
    0.25
    æ¯ı个人çļĦ
    0.25
    -picker
    0.25
    entence
    0.24
    Activations Density 0.010%

    No Known Activations