© 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. 97930
    Prev
    Next
    INDEX
    Explanations

    whose

    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
    bars
    -0.29
    pler
    -0.27
     [('
    -0.26
    åĩ¡æĺ¯
    -0.26
     pisc
    -0.25
     bear
    -0.24
    official
    -0.24
     denen
    -0.24
    ä¸ĢåŃĹ
    -0.24
     AssemblyDescription
    -0.23
    POSITIVE LOGITS
    ãĤĤãģ®ãģ§ãģĻ
    0.28
    è¿ĻäºĽä¸ľè¥¿
    0.25
    ä¼ģä¸ļæıIJä¾Ľ
    0.25
    acl
    0.25
    们éĥ½
    0.24
    _TRA
    0.24
    ç´IJ
    0.24
     Haj
    0.23
    æĹ¥åŃIJ
    0.23
    æŀľ
    0.23
    Activations Density 0.009%

    No Known Activations