© 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. 95442
    Prev
    Next
    INDEX
    Explanations

    Federal

    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
    çi
    -0.29
     fall
    -0.26
     sä
    -0.26
    Capture
    -0.25
     capture
    -0.24
    .mx
    -0.24
    ç§
    -0.24
     age
    -0.24
     falls
    -0.24
    capture
    -0.23
    POSITIVE LOGITS
    æĿ¥çĶµ
    0.31
    宽
    0.28
    StatusBar
    0.28
    æĻ´
    0.27
    洪水
    0.27
     getTitle
    0.27
    æĸ¯çī¹
    0.27
    breadcrumb
    0.26
    ander
    0.26
     Narrow
    0.26
    Activations Density 0.005%

    No Known Activations