© 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. 22-TRANSCODER-HP
    7. 88292
    Prev
    Next
    INDEX
    Explanations

    that

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     yourselves
    -0.28
     ourselves
    -0.26
    eterminate
    -0.26
    (AF
    -0.26
    åIJĦèĩªçļĦ
    -0.25
    èĩªå·±
    -0.25
    éħIJ
    -0.25
    åIJĦèĩª
    -0.25
    æł¹æį®ä¸įåIJĮ
    -0.25
     myself
    -0.25
    POSITIVE LOGITS
    æľĢéĢĤåIJĪ
    0.31
     Byron
    0.27
    æľĢåIJĪéĢĤ
    0.25
    ropic
    0.25
     Ari
    0.25
    PrimaryKey
    0.25
    幺
    0.25
    ะ
    0.24
    kus
    0.24
    红åĪ©
    0.24
    Activations Density 0.007%

    No Known Activations