© 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. 112303
    Prev
    Next
    INDEX
    Explanations

    programming code and licenses

    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
     Princip
    -0.26
    æľ¬èº«å°±æĺ¯
    -0.26
    æĮ«
    -0.25
    .cy
    -0.25
    sym
    -0.25
     dec
    -0.24
     Reading
    -0.24
    fiber
    -0.24
    ⽬
    -0.24
    fef
    -0.23
    POSITIVE LOGITS
    pone
    0.26
     заб
    0.25
    ä¸ĢæŶ
    0.25
    asti
    0.24
    èĨı
    0.24
    ographical
    0.24
    oya
    0.24
     inject
    0.24
    poser
    0.23
     chocolates
    0.23
    Activations Density 0.012%

    No Known Activations