© 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. 151053
    Prev
    Next
    INDEX
    Explanations

    Ratios in scientific texts

    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
    Not in Any Lists

    No Comments

    Negative Logits
     ratios
    -0.62
    Ratio
    -0.61
    ratio
    -0.61
     ratio
    -0.59
     Ratio
    -0.58
    _ratio
    -0.56
    æ¯Ķä¾ĭ
    -0.54
    _RATIO
    -0.52
    çļĦæ¯Ķä¾ĭ
    -0.51
     proportion
    -0.46
    POSITIVE LOGITS
    åŀĽ
    0.28
    ensis
    0.26
    ao
    0.26
    iox
    0.25
    oved
    0.25
    apis
    0.25
    (Locale
    0.25
    apist
    0.24
    å²ģæŶ
    0.24
     dotyczÄħ
    0.24
    Activations Density 0.915%

    No Known Activations