© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 4-TRANSCODER-HP
    7. 156
    Prev
    Next
    INDEX
    Explanations

    quit

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_4.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.4.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     umbrella
    -0.27
    orney
    -0.26
    ()(
    -0.26
    æİı
    -0.26
    uC
    -0.25
     uv
    -0.25
     overse
    -0.25
    æŀ¯
    -0.25
     throughout
    -0.24
    çķĻä¸ĭæĿ¥
    -0.24
    POSITIVE LOGITS
    Ļ¨
    0.28
    çĽĺ
    0.27
    æłĩè®°
    0.26
    blast
    0.25
    ches
    0.25
     shoot
    0.25
     numberOfRows
    0.24
    .square
    0.24
    妨
    0.24
    .LENGTH
    0.23
    Activations Density 0.042%

    No Known Activations