© 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. 25-TRANSCODER-HP
    7. 108546
    Prev
    Next
    INDEX
    Explanations

    per

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     familiar
    -0.29
    ìĶ¬
    -0.26
    ::$
    -0.25
    amiliar
    -0.25
    apter
    -0.24
    rollo
    -0.24
     programmed
    -0.24
     dissoci
    -0.24
    -La
    -0.24
    åIJĦè¡ĮåIJĦ
    -0.23
    POSITIVE LOGITS
     Daily
    0.30
     Campaign
    0.30
    å¹³åĿĩæ°´å¹³
    0.28
    æ¯ı天
    0.28
    Daily
    0.28
     daily
    0.28
    æ¯ı天éĥ½
    0.28
    ä¸Ģ次æĢ§
    0.28
    é¢ijçİĩ
    0.27
    daily
    0.25
    Activations Density 0.980%

    No Known Activations