© 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. 107889
    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_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
    etwork
    -0.29
     working
    -0.29
    -working
    -0.27
    -existing
    -0.27
     avoid
    -0.27
    imations
    -0.26
    -Encoding
    -0.25
    imation
    -0.25
    avoid
    -0.25
     pars
    -0.25
    POSITIVE LOGITS
    fel
    0.27
    ç»ĪçĶŁ
    0.26
    ç»Ī身
    0.26
    ordes
    0.25
    ä¸ĩéĩĮ
    0.25
     newInstance
    0.25
     bergen
    0.24
     breve
    0.24
    .AutoField
    0.24
    æŀ°
    0.24
    Activations Density 12.103%

    No Known Activations