© 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. 17731
    Prev
    Next
    INDEX
    Explanations

    equilibrium/equations

    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
    re
    -0.27
    èݱ
    -0.26
    t
    -0.26
    Emp
    -0.26
     tipped
    -0.25
    annotate
    -0.25
    -le
    -0.25
    adow
    -0.25
    MT
    -0.25
    æģ¢å¤įæŃ£å¸¸
    -0.25
    POSITIVE LOGITS
    cxx
    0.27
    èĭ¦ç¬ij
    0.26
    å¾®ç¬ij
    0.26
    inant
    0.25
    inx
    0.24
    (ver
    0.24
    çijķ
    0.24
    æ°ĶåĢĻ
    0.24
    itant
    0.23
    cretion
    0.23
    Activations Density 0.004%

    No Known Activations