© 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. 31-TRANSCODER-HP
    7. 18412
    Prev
    Next
    INDEX
    Explanations

    violence and abuse

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inic
    -0.30
    hus
    -0.27
     Sin
    -0.27
    åŁł
    -0.27
     Exit
    -0.25
    åį¸
    -0.25
    hall
    -0.25
     Exiting
    -0.24
    交
    -0.24
     hunted
    -0.24
    POSITIVE LOGITS
     FC
    0.27
    ç»Ļä»ĸ们
    0.26
    омеÑĢ
    0.26
    adb
    0.26
    grave
    0.25
    ä¸ī代
    0.25
     past
    0.24
    ä¸įäºĪ
    0.24
    enko
    0.24
     gender
    0.24
    Activations Density 0.786%

    No Known Activations