© 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. 112506
    Prev
    Next
    INDEX
    Explanations

    download

    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
    åĢºæĿĥ
    -0.30
    æł¸
    -0.26
    inness
    -0.26
    æĽ´
    -0.25
    JKLM
    -0.24
    dirs
    -0.24
    ness
    -0.23
    rier
    -0.23
    åĪº
    -0.23
    globals
    -0.23
    POSITIVE LOGITS
     Sax
    0.26
    /response
    0.25
    ä»ħä¾Ľ
    0.24
    ä»¿çľŁ
    0.24
    ãģĮãģĤ
    0.24
    éĽĨåĽ¢èĤ¡ä»½
    0.24
     receipt
    0.23
    è°Īè°Ī
    0.23
    éħįæľī
    0.23
    èĶ»
    0.23
    Activations Density 0.127%

    No Known Activations