© 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. 19717
    Prev
    Next
    INDEX
    Explanations

    Instructions or examples

    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
    igid
    -0.34
     simple
    -0.32
     Simple
    -0.31
     simples
    -0.30
     CROSS
    -0.30
    Simple
    -0.29
    ç®Ģåįķ
    -0.27
    simple
    -0.27
     lá»ĩ
    -0.27
    Cross
    -0.27
    POSITIVE LOGITS
    èĬĭ
    0.27
     pitches
    0.26
    inf
    0.26
    flare
    0.26
    æĹłéĻIJ
    0.25
    åĦ¿å¥³
    0.25
    .Fat
    0.25
    wij
    0.24
    åijĬ
    0.24
    -INF
    0.24
    Activations Density 0.654%

    No Known Activations