© 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. 88716
    Prev
    Next
    INDEX
    Explanations

    attached

    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
    uddenly
    -0.28
    @class
    -0.27
    ä¸ĢåĪ»
    -0.26
    kÅĤad
    -0.25
     sooner
    -0.25
    ä¸Ģçľĭ
    -0.25
    ä¸ĢæĿ¡é¾Ļ
    -0.24
    æĿŀ
    -0.24
    ç»ıéªĮ
    -0.24
    年级
    -0.24
    POSITIVE LOGITS
     flank
    0.28
    ÏĨ
    0.27
     planned
    0.27
    ãģ¡
    0.27
    elib
    0.26
     ÏĨ
    0.26
    efa
    0.25
    è¯į
    0.25
     downside
    0.25
    ut
    0.24
    Activations Density 7.520%

    No Known Activations