© 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. 97621
    Prev
    Next
    INDEX
    Explanations

    research documents

    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
    cken
    -0.26
    åįĹåĮĹ
    -0.24
    _put
    -0.24
    ä¸įéĻIJ
    -0.24
     Ñģпин
    -0.23
    ftime
    -0.23
     Blvd
    -0.23
    .pad
    -0.23
    leet
    -0.22
    markets
    -0.22
    POSITIVE LOGITS
    æľįåĬ¡åijĺ
    0.26
    æľ¬è´¨
    0.24
    icas
    0.24
    fault
    0.24
    undle
    0.24
     shave
    0.23
    WI
    0.23
    pta
    0.23
    缸æ¯Ķ
    0.23
    umin
    0.22
    Activations Density 0.003%

    No Known Activations