© 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. 130000
    Prev
    Next
    INDEX
    Explanations

    Social media platforms

    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
    å®ŀè´¨æĢ§
    -0.29
    ä¸įä»ħ
    -0.27
    spacing
    -0.27
    qualified
    -0.26
     cord
    -0.26
    éĻ¤
    -0.26
    åħĪ
    -0.25
    å¤ļ次
    -0.24
    await
    -0.24
    æ¶Ī
    -0.24
    POSITIVE LOGITS
    nip
    0.30
    Independ
    0.26
    Knowing
    0.25
    arden
    0.24
     segue
    0.24
    åĪĻæĺ¯
    0.24
    å¹´ä¸ŃåĽ½
    0.24
    Keeping
    0.23
    åı¯çŁ¥
    0.23
    fulWidget
    0.23
    Activations Density 2.118%

    No Known Activations