© 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. 24-TRANSCODER-HP
    7. 149925
    Prev
    Next
    INDEX
    Explanations

    Liquids/dripping

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æ³¢åĬ¨
    -0.27
    bars
    -0.25
    -parser
    -0.25
    rat
    -0.25
    è·ĥ
    -0.25
    é¼ł
    -0.24
    ä¸Ĭçľĭ
    -0.24
    cla
    -0.24
    仲
    -0.24
    åijĬè¯ī
    -0.24
    POSITIVE LOGITS
    æ»´æ»´
    0.38
    æ»´
    0.38
    æ·ĭ
    0.35
    æµĩ
    0.35
     gutter
    0.33
     spilled
    0.32
    æ»Ĥ
    0.31
    æ¸į
    0.31
     pouring
    0.31
     dripping
    0.31
    Activations Density 1.257%

    No Known Activations