© 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. 27-TRANSCODER-HP
    7. 7835
    Prev
    Next
    INDEX
    Explanations

    direction words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    aina
    -0.29
    üss
    -0.28
     fully
    -0.28
    èĩ¾
    -0.27
    ely
    -0.27
    ssize
    -0.26
     pressures
    -0.25
    iguous
    -0.25
    ukt
    -0.25
    ipelines
    -0.25
    POSITIVE LOGITS
    orch
    0.31
    %;"
    0.31
    =tmp
    0.28
    ritz
    0.28
    åıĹåĪ°å½±åĵį
    0.27
    çĶº
    0.27
    åį°è±¡
    0.26
     Ahead
    0.26
    ä¸ŃåĽ½æĶ¿åºľ
    0.26
    åIJĥåĪ°
    0.25
    Activations Density 0.002%

    No Known Activations