© 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. 107315
    Prev
    Next
    INDEX
    Explanations

    ad copy, promotional content

    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
     tyranny
    -0.31
    spoken
    -0.26
    å·®åĪ«
    -0.26
    åĭ¤åĬ³
    -0.25
     Sass
    -0.25
     moderated
    -0.25
    ugo
    -0.24
     Cly
    -0.24
    rud
    -0.24
    urdy
    -0.24
    POSITIVE LOGITS
    æĬ¥åijĬæľŁåĨħ
    0.27
    ãĤ±
    0.27
    zia
    0.27
    çIJĩ
    0.26
    éģį
    0.26
    å®Ī
    0.26
    稼
    0.25
    Vent
    0.25
    åĽŁ
    0.25
     defect
    0.25
    Activations Density 13.045%

    No Known Activations