© 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. 30-TRANSCODER-HP
    7. 96171
    Prev
    Next
    INDEX
    Explanations

    Questions snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    onis
    -0.30
    obel
    -0.28
    ä¸ĭæĸ¹
    -0.27
    诱
    -0.27
     hit
    -0.26
    设计åĻ¨
    -0.26
    æ»ij
    -0.26
    abei
    -0.25
     Ferd
    -0.25
    æĭ¾
    -0.24
    POSITIVE LOGITS
     nutritious
    0.27
    æĹłç¼ĺ
    0.25
    å¯ĮåIJ«
    0.25
    ç¼ł
    0.25
     nutrient
    0.25
     maintained
    0.24
    ç§ij
    0.24
    omat
    0.23
    aq
    0.23
     mal
    0.23
    Activations Density 0.883%

    No Known Activations