© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 32-TRANSCODER-HP
    7. 43416
    Prev
    Next
    INDEX
    Explanations

    US government and politics

    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
    Not in Any Lists

    No Comments

    Negative Logits
    cycl
    -0.27
    代
    -0.26
    å¸Ŀçİĭ
    -0.25
    ème
    -0.25
     PK
    -0.24
    ican
    -0.24
    le
    -0.23
    æ¸ł
    -0.23
    רצ×Ļ
    -0.23
    _mut
    -0.23
    POSITIVE LOGITS
     White
    0.64
     Administration
    0.59
    White
    0.54
     Secret
    0.51
     Department
    0.50
     administration
    0.49
    Administration
    0.48
    çĻ½
    0.44
    Department
    0.44
     Secretary
    0.44
    Activations Density 0.013%

    No Known Activations