© 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. 107692
    Prev
    Next
    INDEX
    Explanations

    Technical/computational text

    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
    å¹»
    -0.27
     speculation
    -0.25
    ged
    -0.25
    å°ijæķ°
    -0.24
    msgid
    -0.24
    æ²Ļ
    -0.24
    åĮı
    -0.24
    è¿IJ
    -0.24
    åıªæĺ¯ä¸ºäºĨ
    -0.24
    ä¸Ģ个
    -0.23
    POSITIVE LOGITS
    FML
    0.29
    erve
    0.27
    ema
    0.26
    iona
    0.25
    emax
    0.24
    nia
    0.24
    :num
    0.24
    Fed
    0.24
    vvm
    0.24
    ',[
    0.24
    Activations Density 27.087%

    No Known Activations