© 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. 96223
    Prev
    Next
    INDEX
    Explanations

    code

    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
    ละ
    -0.28
    iable
    -0.27
    æĹ¥æĬ¥ç¤¾
    -0.27
    æµģæ·Į
    -0.25
     unsubscribe
    -0.25
    -dismiss
    -0.25
     Forgot
    -0.24
    /article
    -0.24
    Paginator
    -0.24
    MAIL
    -0.24
    POSITIVE LOGITS
    OOD
    0.26
    dee
    0.26
     roy
    0.26
    aic
    0.26
    è¿«
    0.25
     witnesses
    0.25
     ansch
    0.25
    à¹Ĥฮ
    0.24
     deep
    0.24
    ér
    0.24
    Activations Density 0.882%

    No Known Activations