© 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. 31-TRANSCODER-HP
    7. 18763
    Prev
    Next
    INDEX
    Explanations

    amounts and sets

    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
    crast
    -0.28
     a
    -0.28
    umper
    -0.28
    è¾ŀ
    -0.26
    ypse
    -0.26
    +a
    -0.26
    æ±
    -0.26
    ä¸įæĪIJ
    -0.25
    oes
    -0.25
    åı¥
    -0.24
    POSITIVE LOGITS
     fixed
    0.27
     Fixed
    0.26
     Sund
    0.26
    OutOf
    0.25
    whole
    0.25
    fixed
    0.25
     buz
    0.24
     FIXED
    0.24
    Fixed
    0.24
     fucking
    0.23
    Activations Density 0.230%

    No Known Activations