© 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. 19-TRANSCODER-HP
    7. 9097
    Prev
    Next
    INDEX
    Explanations

    News/announcements

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     consume
    -0.27
    roken
    -0.25
    纲
    -0.24
    åĥŃ
    -0.24
    课åłĤæķĻåѦ
    -0.24
    åı¯èĥ½åŃĺåľ¨
    -0.24
    饮çĶ¨
    -0.24
    å¹½
    -0.24
    tering
    -0.23
    splash
    -0.23
    POSITIVE LOGITS
     FTC
    0.25
     bem
    0.24
    åħ³åģľ
    0.24
    REATE
    0.23
     widespread
    0.23
     invasive
    0.23
    对ä¸ŃåĽ½
    0.23
     universal
    0.23
    Decorator
    0.23
     entend
    0.23
    Activations Density 0.053%

    No Known Activations