© 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. 149497
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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
    -doc
    -0.30
    <![
    -0.26
     yourselves
    -0.26
    è§Ī
    -0.25
    ë²Į
    -0.25
    è¿Ļæīįæĺ¯
    -0.24
    viewport
    -0.24
    ves
    -0.24
    appl
    -0.23
    zeug
    -0.23
    POSITIVE LOGITS
    éĢļ常æĺ¯
    1.06
     typically
    0.97
     usually
    0.95
    éĢļ常
    0.85
    ä¸ĢèĪ¬æĺ¯
    0.81
     Typically
    0.80
     Usually
    0.77
    typically
    0.76
     probably
    0.73
    usually
    0.73
    Activations Density 0.978%

    No Known Activations