© 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. 33-TRANSCODER-HP
    7. 53699
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    çļĦæł¸å¿ĥ
    -0.33
    æł¸å¿ĥ
    -0.31
     Corey
    -0.29
    otle
    -0.29
    cid
    -0.28
    ä¹ĭæĺŁ
    -0.27
    plusplus
    -0.27
    Core
    -0.27
    ]++;↵
    -0.27
    (Core
    -0.27
    POSITIVE LOGITS
     cutting
    0.47
     under
    0.41
     cut
    0.39
     CUT
    0.35
     Cutting
    0.33
    åĪĩåī²
    0.32
     cuts
    0.32
    åīª
    0.31
    -cut
    0.31
     inside
    0.30
    Activations Density 0.011%

    No Known Activations