© 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. 129889
    Prev
    Next
    INDEX
    Explanations

    (

    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
    SelectionMode
    -0.29
    ULAR
    -0.27
     curs
    -0.26
    æĮĩåIJij
    -0.26
    favor
    -0.25
    åħ·ä½ĵçļĦ
    -0.25
    specific
    -0.25
     Ptr
    -0.24
     offensive
    -0.24
    pur
    -0.24
    POSITIVE LOGITS
    loh
    0.29
    ipsoid
    0.28
    lags
    0.27
    èĢĮåıĪ
    0.27
    -spin
    0.27
    lop
    0.26
     ConfigurationManager
    0.26
    åħ¶æīĢ
    0.25
    ies
    0.25
    离å¼Ģ
    0.25
    Activations Density 5.046%

    No Known Activations