© 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. 30-TRANSCODER-HP
    7. 96768
    Prev
    Next
    INDEX
    Explanations

    the

    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
    +p
    -0.28
    离åİ»
    -0.26
    .FLAG
    -0.26
    æķķ
    -0.25
     rationale
    -0.25
    cone
    -0.24
    æĭ¨
    -0.24
    讽
    -0.24
    $",
    -0.24
    _unset
    -0.23
    POSITIVE LOGITS
    该åİ¿
    0.32
    ä¹ĭåŁİ
    0.31
    å±±èĦī
    0.30
     cities
    0.29
     valley
    0.29
    åŁİå¸Ĥ
    0.29
    å±±åİ¿
    0.29
     lands
    0.28
     sands
    0.28
     mountains
    0.27
    Activations Density 0.886%

    No Known Activations