© 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. 35-TRANSCODER-HP
    7. 38738
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æ°´éĿ¢
    -0.27
    å¹Ķ
    -0.27
     perf
    -0.25
     selectedIndex
    -0.25
    üss
    -0.25
     humiliating
    -0.25
    åĮ£
    -0.24
    è¾ĸ
    -0.24
    indx
    -0.24
    æļĤ
    -0.24
    POSITIVE LOGITS
    geries
    0.28
     Richt
    0.28
    okit
    0.27
    .Call
    0.26
     Crash
    0.26
    FU
    0.25
    ¢
    0.25
    tons
    0.25
    踬
    0.25
    æĹ©
    0.25
    Activations Density 4.407%

    No Known Activations

    This feature has no known activations.