© 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. 17926
    Prev
    Next
    INDEX
    Explanations

    place names and proper nouns

    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
    ìĦ¸
    -0.27
    opher
    -0.26
    trib
    -0.25
    rons
    -0.25
     Loose
    -0.25
    atan
    -0.24
     (${
    -0.24
    æ²»çIJĨä½ĵç³»
    -0.23
     thought
    -0.23
    arrass
    -0.23
    POSITIVE LOGITS
    å°º
    0.27
     Isl
    0.26
    unt
    0.26
    imed
    0.26
    åºľ
    0.26
    alg
    0.25
    æ»ij
    0.25
    OwnProperty
    0.25
    allet
    0.25
    ippi
    0.25
    Activations Density 39.546%

    No Known Activations