© 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. 32-TRANSCODER-HP
    7. 42817
    Prev
    Next
    INDEX
    Explanations

    say from

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å°Ĩ以
    -0.33
    between
    -0.29
    import
    -0.29
     import
    -0.28
    Between
    -0.27
    åĩºæĿ¥çļĦ
    -0.27
     Between
    -0.27
    ä½łåľ¨
    -0.26
    èĢĮåľ¨
    -0.26
    åĭ¢
    -0.26
    POSITIVE LOGITS
    rom
    0.59
     rom
    0.57
    ROM
    0.57
    _rom
    0.51
     ROM
    0.51
    ä»İä¸ļ
    0.50
     Rom
    0.46
    ä»İæľª
    0.45
     fram
    0.43
     trom
    0.42
    Activations Density 0.031%

    No Known Activations