© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 24-TRANSCODER-HP
    7. 151179
    Prev
    Next
    INDEX
    Explanations

    Code and non-English snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     bree
    -0.28
    è¯ķè¡Į
    -0.26
    atest
    -0.26
    ĥģ
    -0.25
    ertest
    -0.25
     son
    -0.25
     commissioned
    -0.25
     generally
    -0.25
    à¹Ģศ
    -0.25
     testing
    -0.24
    POSITIVE LOGITS
    è¿ĶåĽŀæIJľçĭIJ
    0.31
    .dtd
    0.27
     —↵↵
    0.27
    èĭ¦
    0.27
    åı¥
    0.26
    æĵ¦
    0.26
    åħ¬å¹³
    0.26
    (routes
    0.25
    People
    0.25
    MET
    0.25
    Activations Density 0.044%

    No Known Activations