© 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. 62969
    Prev
    Next
    INDEX
    Explanations

    hyphen

    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
    åIJĦèĩª
    -0.27
    uner
    -0.27
    ateg
    -0.27
    achment
    -0.26
    çIJŃ
    -0.25
    RE
    -0.25
    åIJĦèĩªçļĦ
    -0.24
    asers
    -0.24
    Anal
    -0.24
    .quest
    -0.24
    POSITIVE LOGITS
    èĦĸ
    0.28
    Äĩi
    0.24
    fila
    0.23
    elen
    0.23
    ترا
    0.23
     liebe
    0.23
    èIJĭ
    0.23
     sprung
    0.23
    Vu
    0.23
    _subset
    0.23
    Activations Density 13.903%

    No Known Activations