© 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. 96623
    Prev
    Next
    INDEX
    Explanations

    Code and SQL-related content

    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
    åŃ£
    -0.29
     season
    -0.28
     seasonal
    -0.28
    season
    -0.27
    åŃ£èĬĤ
    -0.26
    _season
    -0.26
    ÑĩÑĮ
    -0.26
    tl
    -0.26
    ATOM
    -0.24
     specialized
    -0.24
    POSITIVE LOGITS
    æĵį
    0.32
    æĩĭ
    0.29
    ä¿Ŀå®ī
    0.28
    å·®ä¸įå¤ļ
    0.28
    ä¹ĭåĨħ
    0.27
    ä¸į说
    0.27
    åģľæ»ŀ
    0.27
     Murdoch
    0.27
    -pre
    0.27
    æĵįä½ľ
    0.27
    Activations Density 0.100%

    No Known Activations