© 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. 20-TRANSCODER-HP
    7. 108154
    Prev
    Next
    INDEX
    Explanations

    computer code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æį¢æĿ¥
    -0.28
    第ä¸Ģå±Ĭ
    -0.27
    oad
    -0.27
    rush
    -0.27
    sut
    -0.26
    dana
    -0.26
    åīįæīĢæľªæľī
    -0.26
    æľIJ
    -0.26
    roys
    -0.26
    hydr
    -0.26
    POSITIVE LOGITS
    页
    0.30
    æijĺ
    0.27
     blij
    0.27
    _pg
    0.25
     pg
    0.25
     let
    0.24
    CC
    0.24
     f
    0.24
    .getSimpleName
    0.24
    Normalization
    0.24
    Activations Density 3.597%

    No Known Activations