© 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. 107021
    Prev
    Next
    INDEX
    Explanations

    General conversational text

    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
    Ĺı
    -0.29
    ichel
    -0.28
    esub
    -0.27
     negate
    -0.25
    )!↵
    -0.24
    ä»ĸå·²ç»ı
    -0.24
    è®®
    -0.24
    /null
    -0.23
    åŁİ乡
    -0.23
    éĢ¼è¿ij
    -0.23
    POSITIVE LOGITS
    åłª
    0.27
    è´¹çĶ¨
    0.25
    abyrinth
    0.24
    éĢĢç¨İ
    0.24
     afl
    0.24
    problems
    0.24
    nc
    0.24
     unpopular
    0.24
     ensuing
    0.23
    umer
    0.23
    Activations Density 13.687%

    No Known Activations