© 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. 95769
    Prev
    Next
    INDEX
    Explanations

    Iraq and Baghdad

    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
    edException
    -0.30
    eries
    -0.25
     Cyril
    -0.24
    ÙĬدÙĬ
    -0.24
    isman
    -0.24
    äºĮåŃĹ
    -0.23
    çłĶ究éĻ¢
    -0.23
     fulfilled
    -0.23
    以èĩ³äºİ
    -0.23
    åĪĿ
    -0.23
    POSITIVE LOGITS
    NewItem
    0.27
    æļij
    0.26
    èį«
    0.25
    追éĹ®
    0.25
     Cave
    0.24
    ::::::::
    0.24
    chemes
    0.24
    tempts
    0.24
     flex
    0.24
     axis
    0.23
    Activations Density 0.107%

    No Known Activations