© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 35-TRANSCODER-HP
    7. 39634
    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
    Not in Any Lists

    No Comments

    Negative Logits
    åĮĹ京æŶéĹ´
    -0.27
    çIJ¶
    -0.25
    é«ĺæķĪçļĦ
    -0.25
     militia
    -0.23
     gấp
    -0.23
     conscience
    -0.23
    åĩŃ
    -0.23
    Inside
    -0.23
    oral
    -0.22
     mát
    -0.22
    POSITIVE LOGITS
     LENG
    0.27
    平稳
    0.26
    åľ¨è·¯ä¸Ĭ
    0.26
    è¿Ļæĸ¹éĿ¢
    0.26
    eyer
    0.25
    bove
    0.25
    bsites
    0.25
    uten
    0.24
    ç¢
    0.24
    obra
    0.23
    Activations Density 0.207%

    No Known Activations

    This feature has no known activations.