© 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. 31-TRANSCODER-HP
    7. 19847
    Prev
    Next
    INDEX
    Explanations

    for

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    åŀĭ
    -0.26
     revolves
    -0.24
    uga
    -0.24
    enso
    -0.24
    è¡Įåĭķ
    -0.24
     highlights
    -0.24
     haystack
    -0.23
    å¼ı
    -0.23
    è¡ĮåĬ¨
    -0.23
     hé
    -0.23
    POSITIVE LOGITS
     for
    0.76
    for
    0.67
    	for
    0.55
     длÑı
    0.52
    _for
    0.47
     FOR
    0.47
    For
    0.47
    对äºİ
    0.45
     für
    0.45
    ForKey
    0.45
    Activations Density 4.070%

    No Known Activations