© 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. 43540
    Prev
    Next
    INDEX
    Explanations

    components or factors

    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
    ä¸ĵ访
    -0.27
    ç¾¤å²Ľ
    -0.25
    æĬ¥ä¸ļ
    -0.25
    ç¼ľ
    -0.25
    ä¸Ģæľ¬ä¹¦
    -0.25
    è¡ĮæĶ¿åĮº
    -0.24
    okies
    -0.24
    æĽ²æĬĺ
    -0.24
     onslaught
    -0.24
    ä¸Ģæł¹
    -0.24
    POSITIVE LOGITS
     to
    0.45
     central
    0.40
     driving
    0.39
     limiting
    0.38
     supporting
    0.37
     requiring
    0.35
     contributing
    0.35
     that
    0.35
    Central
    0.34
     drivers
    0.33
    Activations Density 0.007%

    No Known Activations