© 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. 32-TRANSCODER-HP
    7. 44025
    Prev
    Next
    INDEX
    Explanations

    (

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     pó
    -0.29
     ot
    -0.28
    _pb
    -0.27
     mrb
    -0.26
    озв
    -0.26
    xbc
    -0.26
    btc
    -0.25
    çIJ¥
    -0.25
    âĢ¦)↵↵
    -0.25
    llib
    -0.25
    POSITIVE LOGITS
    çīµ
    0.31
    æĿł
    0.30
    _FLAG
    0.28
    çݯ
    0.27
    èĮĥåĽ´
    0.26
    çīµå¼ķ
    0.25
    èģĶ绾
    0.25
    ä¿¡åı·
    0.25
    è´¢
    0.25
     initialState
    0.25
    Activations Density 0.010%

    No Known Activations