© 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. 107181
    Prev
    Next
    INDEX
    Explanations

    Weapons

    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
    ç½ĺ
    -0.28
    би
    -0.28
    yı
    -0.27
    åħ¬åı¸ç«łç¨ĭ
    -0.27
    åı¦è¡Į
    -0.25
    å·®
    -0.25
    \/\/
    -0.25
    oko
    -0.24
     Patent
    -0.24
    arie
    -0.24
    POSITIVE LOGITS
     dangerous
    0.27
    seed
    0.26
    iker
    0.26
    æ´Ĺ
    0.26
    worth
    0.25
    knife
    0.24
     bundle
    0.24
    licht
    0.24
     tick
    0.24
     Beats
    0.24
    Activations Density 7.113%

    No Known Activations