© 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. 18983
    Prev
    Next
    INDEX
    Explanations

    capitalized abbreviations

    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
    æĭ¥æĬ¤
    -0.28
    ä¹Į
    -0.27
    OTO
    -0.27
    ç¥Ŀè´º
    -0.27
    黼
    -0.26
    paration
    -0.25
    亲
    -0.24
    çIJ¥
    -0.24
    ÑĪе
    -0.24
    oint
    -0.24
    POSITIVE LOGITS
    ä¸ļ主
    0.30
    _ve
    0.25
    bre
    0.25
    ASC
    0.25
    yms
    0.25
    ç¼ł
    0.24
    ych
    0.24
    èĭŁ
    0.24
    ivals
    0.24
    Owners
    0.24
    Activations Density 0.978%

    No Known Activations