Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Qwen3-4B
    3. 23-TRANSCODER-HP
    4. 7475
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_23.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.23.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Âł
    -0.29
    chema
    -0.28
    olib
    -0.27
    iola
    -0.27
    æĬ
    -0.27
    ereum
    -0.24
    }->{
    -0.24
    åĭĻ
    -0.24
    \xa
    -0.24
    itude
    -0.24
    POSITIVE LOGITS
     Field
    0.29
    æľªæĿ¥çļĦ
    0.26
     FIELD
    0.26
    .Field
    0.25
    åIJĪéĢĤ
    0.25
     Effective
    0.25
    ingham
    0.24
     Numerous
    0.23
    ìĽ¨
    0.23
    iso
    0.23
    Activations Density 0.007%

    No Known Activations