© 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. 95930
    Prev
    Next
    INDEX
    Explanations

    leopard

    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
     living
    -0.33
     Living
    -0.33
    æ´»
    -0.31
    ç¡IJ
    -0.31
     Opening
    -0.29
    æĶ¯æĴij
    -0.29
    å¼Ģåı£
    -0.29
     Open
    -0.29
    主
    -0.29
     opening
    -0.28
    POSITIVE LOGITS
    æĥķ
    0.28
    indre
    0.27
    üs
    0.26
    åIJĦåĮº
    0.25
    crawler
    0.25
    硬åĮĸ
    0.25
    vent
    0.25
    åĽ¢åijĺ
    0.25
    clus
    0.24
    Https
    0.24
    Activations Density 0.104%

    No Known Activations