© 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. 22-TRANSCODER-HP
    7. 98529
    Prev
    Next
    INDEX
    Explanations

    Code symbols

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ç´¯äºĨ
    -0.26
     isError
    -0.26
    导åħ¥
    -0.25
    ç³»
    -0.25
    cow
    -0.25
     underestimated
    -0.24
    åħľ
    -0.24
    èļ¬
    -0.24
    åħ¨æĸ¹ä½į
    -0.24
     EVERY
    -0.23
    POSITIVE LOGITS
    -season
    0.27
    .stage
    0.24
    -bed
    0.24
    Summer
    0.24
    paragraph
    0.24
     Rape
    0.24
     Remarks
    0.24
    ioned
    0.24
    azor
    0.23
    inclusive
    0.23
    Activations Density 0.014%

    No Known Activations