© 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. 44111
    Prev
    Next
    INDEX
    Explanations

    code and configuration

    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
     close
    -0.33
    /close
    -0.32
    	close
    -0.29
    light
    -0.28
    close
    -0.28
    adj
    -0.28
    iles
    -0.27
    iled
    -0.27
    heat
    -0.27
    ite
    -0.27
    POSITIVE LOGITS
     reference
    0.32
     interfaces
    0.31
    reference
    0.31
    _interfaces
    0.30
     interface
    0.29
    æİ¥åı£
    0.29
     Interfaces
    0.29
    ä¸įå®Į
    0.27
    otas
    0.27
     house
    0.27
    Activations Density 0.001%

    No Known Activations