© 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. 17977
    Prev
    Next
    INDEX
    Explanations

    bad

    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
     good
    -0.64
     Good
    -0.58
    好
    -0.57
    good
    -0.54
    好çļĦ
    -0.54
    Good
    -0.52
     GOOD
    -0.50
    -good
    -0.41
     poor
    -0.41
    .good
    -0.40
    POSITIVE LOGITS
    bling
    0.30
    LING
    0.29
    PTION
    0.28
    çĮ·
    0.27
    enda
    0.26
    _nil
    0.26
    ENO
    0.26
    ienia
    0.25
    -backed
    0.25
    Nil
    0.25
    Activations Density 0.013%

    No Known Activations