© 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. 108236
    Prev
    Next
    INDEX
    Explanations

    medicine

    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
    tlement
    -0.26
    component
    -0.26
    .perform
    -0.25
    adera
    -0.24
    tur
    -0.24
    URNS
    -0.24
    è¡¡
    -0.24
    身ä½ĵåģ¥åº·
    -0.24
    ä¸ľè¥¿
    -0.24
    LiveData
    -0.24
    POSITIVE LOGITS
    éĹ´éļĻ
    0.27
    Interop
    0.26
    vet
    0.26
    agas
    0.26
    issement
    0.25
    wij
    0.25
    iji
    0.25
    eps
    0.24
    urry
    0.24
    erde
    0.24
    Activations Density 37.244%

    No Known Activations