© 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. 17971
    Prev
    Next
    INDEX
    Explanations

    non-English text

    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
    åĢŁæ¬¾äºº
    -0.27
    innie
    -0.24
    boxed
    -0.24
    天涯
    -0.24
    å»Ĭ
    -0.23
    èĬª
    -0.23
    /host
    -0.23
     ausge
    -0.23
    uspended
    -0.23
    .tight
    -0.23
    POSITIVE LOGITS
    åģļçļĦ
    0.92
    åģļ
    0.87
    åģļäºĨ
    0.81
    åģļåĩº
    0.72
     doing
    0.71
     done
    0.69
     Doing
    0.66
    è¦ģåģļ
    0.66
     lÃłm
    0.66
    åģļå¾Ĺ
    0.65
    Activations Density 1.620%

    No Known Activations