© 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. 99059
    Prev
    Next
    INDEX
    Explanations

    increase

    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
    don
    -0.27
    迹
    -0.26
    ä¼´ä¾£
    -0.26
    æĪĸå¤ļæĪĸå°ij
    -0.24
     AHL
    -0.24
    icide
    -0.24
    yla
    -0.24
    arsed
    -0.24
    --↵↵
    -0.24
    RY
    -0.24
    POSITIVE LOGITS
    å±Ī
    0.35
     vulnerability
    0.31
    ä¸įåĪ©
    0.31
    宣
    0.28
     vulnerabilities
    0.26
    Ñĥв
    0.26
    忧èĻij
    0.25
    _workers
    0.25
    (worker
    0.25
    emes
    0.25
    Activations Density 0.004%

    No Known Activations