© 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. 18-TRANSCODER-HP
    7. 89882
    Prev
    Next
    INDEX
    Explanations

    8

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    åŁķ
    -0.31
    åķ¡
    -0.27
    ObjectName
    -0.26
    第ä¸ĢèĬĤ
    -0.25
    /unit
    -0.25
    åĸī
    -0.24
    以ä¸ĬåѦåİĨ
    -0.24
     exposition
    -0.24
     comprend
    -0.23
    zym
    -0.23
    POSITIVE LOGITS
    __(
    0.27
    ä¸ĩäºĭ
    0.25
    setQuery
    0.24
    éħįåIJĪ
    0.24
    汪
    0.24
    erval
    0.24
    åģ¥åº·
    0.24
    based
    0.23
    obot
    0.23
    lycer
    0.23
    Activations Density 4.405%

    No Known Activations