© 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. 63704
    Prev
    Next
    INDEX
    Explanations

    register

    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
    åĩ¡äºĭ
    -0.29
    åı¯èĥ½æĢ§
    -0.27
     <![
    -0.27
    ativ
    -0.26
    eries
    -0.25
    åijĬè¯īä½ł
    -0.24
    ç¾İ人
    -0.24
     substrate
    -0.24
    -tip
    -0.24
     establishment
    -0.24
    POSITIVE LOGITS
    Ŀå§ĭ
    0.27
    cing
    0.27
     Ansi
    0.26
    ç³Ļ
    0.26
    vro
    0.25
     resolves
    0.24
    -separated
    0.24
    æ¥ŀ
    0.24
    .defaults
    0.23
    resolved
    0.23
    Activations Density 0.347%

    No Known Activations