© 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. 87208
    Prev
    Next
    INDEX
    Explanations

    email

    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
     numbering
    -0.29
     stroke
    -0.27
    å¹³
    -0.25
     species
    -0.25
     flat
    -0.25
     finally
    -0.24
    ENCIES
    -0.24
     number
    -0.24
    Pe
    -0.24
     num
    -0.24
    POSITIVE LOGITS
    mods
    0.29
    çļĦæ¨¡æł·
    0.26
    versions
    0.25
    SetName
    0.25
     spur
    0.25
    å¥Ĺè£ħ
    0.24
    SPARENT
    0.24
     OTHERWISE
    0.23
    ters
    0.23
    /archive
    0.23
    Activations Density 0.005%

    No Known Activations