© 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. 18-TRANSCODER-HP
    7. 88730
    Prev
    Next
    INDEX
    Explanations

    research and social issues

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    elihood
    -0.33
    vero
    -0.26
    anders
    -0.25
    ãģ¶ãĤĬ
    -0.25
    æį§
    -0.25
    èĤļåŃIJ
    -0.24
    åł¡åŀĴ
    -0.24
    aat
    -0.24
     localVar
    -0.24
    ports
    -0.24
    POSITIVE LOGITS
    ideo
    0.28
     iT
    0.27
     Dane
    0.27
     inter
    0.25
    -dat
    0.25
    inte
    0.24
    edd
    0.24
    éĥ½ä¸įæĺ¯
    0.24
    Address
    0.24
    ä¹Łåı¯
    0.23
    Activations Density 27.380%

    No Known Activations