© 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. 25-TRANSCODER-HP
    7. 106986
    Prev
    Next
    INDEX
    Explanations

    code/mathematical notation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    oes
    -0.28
    湿
    -0.28
    è²Į
    -0.27
     wild
    -0.26
    åİ¿éķ¿
    -0.26
     Crafts
    -0.25
    ields
    -0.25
     mapping
    -0.25
    (mapping
    -0.25
    wild
    -0.25
    POSITIVE LOGITS
    rophe
    0.25
    aptcha
    0.25
    ammen
    0.25
    entar
    0.24
    fir
    0.24
    orer
    0.24
     dus
    0.24
    zte
    0.24
    incare
    0.23
    achines
    0.23
    Activations Density 1.009%

    No Known Activations