© 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. 98728
    Prev
    Next
    INDEX
    Explanations

    Providing information

    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
     triangular
    -0.29
    asm
    -0.28
     Held
    -0.27
    欺è¯Ī
    -0.27
    dba
    -0.25
    ä¸Ńå¤ĸ
    -0.25
    ä¸īè§Ĵ
    -0.25
    Women
    -0.25
     fresh
    -0.24
    å°ı
    -0.24
    POSITIVE LOGITS
    rokes
    0.29
    gings
    0.28
     cáo
    0.28
    isen
    0.26
     Cheer
    0.25
    èµĽåIJİ
    0.25
    -addons
    0.25
     RemoteException
    0.25
     convin
    0.24
    IJľç´¢
    0.24
    Activations Density 0.004%

    No Known Activations