© 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. 88520
    Prev
    Next
    INDEX
    Explanations

    numbers

    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
    ibel
    -0.30
    PFN
    -0.29
     ÄIJình
    -0.28
    elden
    -0.28
    ç¥ŀå·ŀ
    -0.28
    rollers
    -0.25
     Mango
    -0.25
    otherwise
    -0.24
    ADM
    -0.24
    respect
    -0.24
    POSITIVE LOGITS
     relative
    0.28
     evt
    0.27
     alt
    0.26
    æľīçĶ¨çļĦ
    0.26
     spans
    0.26
     EVERY
    0.25
    æĹ©ä¸Ĭ
    0.24
    (relative
    0.24
     dif
    0.24
    diff
    0.24
    Activations Density 0.006%

    No Known Activations