© 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. 27-TRANSCODER-HP
    7. 9617
    Prev
    Next
    INDEX
    Explanations

    Code or mixed content

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æľĢç»Ī
    -0.93
     final
    -0.89
    æľĢçµĤ
    -0.77
    final
    -0.75
     FINAL
    -0.69
     Final
    -0.68
    	final
    -0.68
     ultimately
    -0.68
    Final
    -0.67
    -final
    -0.67
    POSITIVE LOGITS
     careful
    0.25
    仪å¼ı
    0.25
     carefully
    0.25
     %#
    0.25
     operation
    0.25
    Explicit
    0.25
    éré
    0.25
    idth
    0.24
     Vand
    0.24
    Quiet
    0.24
    Activations Density 0.016%

    No Known Activations