© 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. 63180
    Prev
    Next
    INDEX
    Explanations

    coding and technical discussions

    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
     accordingly
    -0.31
    âĢ¢↵↵
    -0.31
     ultimately
    -0.31
     precisely
    -0.30
     consequently
    -0.30
     literally
    -0.28
     furthermore
    -0.28
    åīįæīĢ
    -0.27
     subsequently
    -0.27
     fortunately
    -0.26
    POSITIVE LOGITS
    ene
    0.30
    -abs
    0.27
    èĴĻ
    0.25
    èł¡
    0.25
     tipping
    0.24
    ikk
    0.24
    VK
    0.24
    kl
    0.23
    è¾Ķ
    0.23
    ĮĢ
    0.23
    Activations Density 60.672%

    No Known Activations