© 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. 112606
    Prev
    Next
    INDEX
    Explanations

    code and boolean logic

    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
    å¯Ĵ
    -0.29
    责任人
    -0.28
    cente
    -0.25
    ALK
    -0.25
    vara
    -0.25
    æŶéĹ´åĴĮ
    -0.24
    emes
    -0.24
    utive
    -0.24
    _transient
    -0.24
    duct
    -0.23
    POSITIVE LOGITS
     gì
    0.26
    éĩİ
    0.25
    çĶļèĩ³æĺ¯
    0.25
    /validation
    0.24
     Carlson
    0.23
    .effect
    0.23
     literally
    0.23
    ä¸įåı¯æĢĿ
    0.23
    Sock
    0.22
    	answer
    0.22
    Activations Density 0.003%

    No Known Activations