© 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. 30-TRANSCODER-HP
    7. 95105
    Prev
    Next
    INDEX
    Explanations

    academic papers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æĸĩæľ¬
    -0.29
     miscellaneous
    -0.28
    xFFFFFFFF
    -0.27
    ertext
    -0.26
     закон
    -0.26
    resher
    -0.26
    aze
    -0.26
     text
    -0.26
     creeping
    -0.25
    æľ¦èĥ§
    -0.25
    POSITIVE LOGITS
    èħ¹
    0.34
    èħ°
    0.31
    èħ°éĥ¨
    0.30
    å¼ĥ
    0.28
     Successfully
    0.26
    é¢Ī
    0.26
    /respond
    0.25
    tur
    0.25
    èĭı
    0.25
    Respond
    0.24
    Activations Density 0.783%

    No Known Activations