© 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. 129711
    Prev
    Next
    INDEX
    Explanations

    News articles/opinion writing

    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
    æijĨ
    -0.31
    IONS
    -0.30
    parts
    -0.29
    åij¼å£°
    -0.28
     Sunday
    -0.26
     titular
    -0.26
    пÑĥ
    -0.25
    åıij
    -0.25
    kont
    -0.24
    itä
    -0.24
    POSITIVE LOGITS
     Anthem
    0.27
     BaseActivity
    0.26
     compatible
    0.26
    ມ
    0.26
    第ä¸Ģ次
    0.25
    brtc
    0.25
    åľ¨ä¸ŃåĽ½
    0.25
     vaguely
    0.25
    uba
    0.24
    bf
    0.23
    Activations Density 11.128%

    No Known Activations