© 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. 130679
    Prev
    Next
    INDEX
    Explanations

    Internet forum discussions

    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
     inade
    -0.37
     shortcomings
    -0.35
    缺ä¹ı
    -0.35
     inadequate
    -0.33
    fail
    -0.33
     fails
    -0.33
     Fail
    -0.32
    æ¬łç¼º
    -0.31
    ä¼ĺè¶Ĭ
    -0.31
     fail
    -0.31
    POSITIVE LOGITS
    æľºä¼ļ
    0.25
    åŁºæķ°
    0.24
    .writer
    0.24
    .iter
    0.23
    ëĭIJ
    0.23
     teen
    0.23
    itre
    0.23
    recht
    0.23
     phen
    0.23
     preserved
    0.23
    Activations Density 49.526%

    No Known Activations