© 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. 18655
    Prev
    Next
    INDEX
    Explanations

    notifications and messages

    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
    éĢļçŁ¥
    -1.16
     notification
    -1.16
     notify
    -1.09
     notifications
    -1.09
     Notification
    -1.08
    notify
    -1.06
    notification
    -1.05
    Notification
    -1.04
    çļĦéĢļçŁ¥
    -1.03
     Notify
    -0.98
    POSITIVE LOGITS
    LU
    0.27
    kes
    0.27
    rng
    0.26
    ernals
    0.24
    hya
    0.24
    Translated
    0.23
    éªĽ
    0.23
    olics
    0.23
    Lua
    0.23
    лÑİ
    0.23
    Activations Density 1.455%

    No Known Activations