© 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. 30-TRANSCODER-HP
    7. 96029
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unfold
    -0.29
     transcript
    -0.28
    èĬ
    -0.27
    ãĤįãģĨ
    -0.27
    ansen
    -0.27
     accounts
    -0.26
    åIJĽ
    -0.26
     feed
    -0.26
     simply
    -0.26
     Accounts
    -0.25
    POSITIVE LOGITS
    æĤ¬æµ®
    0.29
    æĿ¯
    0.28
    ç´¯äºĨ
    0.28
    æĿ¯åŃIJ
    0.28
    ÑģÑĤÑĢел
    0.27
    elter
    0.27
    ampion
    0.26
    hole
    0.26
     suspended
    0.25
    lotte
    0.25
    Activations Density 0.096%

    No Known Activations