© 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. 111665
    Prev
    Next
    INDEX
    Explanations

    reports

    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
     aren
    -0.26
    ä¸įæĺ¯
    -0.25
     Trades
    -0.25
     wasn
    -0.24
    [array
    -0.24
    软
    -0.24
     weren
    -0.24
     stays
    -0.24
    夥
    -0.23
    åıĭ
    -0.23
    POSITIVE LOGITS
    è¿Ľä¸ĢæŃ¥
    0.28
     Äijá»Ļc
    0.27
    grund
    0.27
    ards
    0.27
    /GL
    0.27
     ded
    0.26
    çĭ¬
    0.25
    uni
    0.25
    éĢ²ä¸ĢæŃ¥
    0.25
    çį¨
    0.25
    Activations Density 0.004%

    No Known Activations