© 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. 18068
    Prev
    Next
    INDEX
    Explanations

    and

    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
    Not in Any Lists

    No Comments

    Negative Logits
     about
    -0.29
    Ub
    -0.25
    Unmarshaller
    -0.25
     About
    -0.24
     ek
    -0.24
    Exporter
    -0.24
    about
    -0.24
    undra
    -0.23
    ittal
    -0.23
     expand
    -0.23
    POSITIVE LOGITS
    æīĺ
    0.28
    åij¼
    0.28
    ä¸ĥ
    0.26
    æīĭèħķ
    0.26
    éĢ»è¾ij
    0.25
    esian
    0.25
    orce
    0.25
    eric
    0.25
    èĩªçĦ¶
    0.25
     Shiv
    0.24
    Activations Density 0.154%

    No Known Activations