© 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. 18828
    Prev
    Next
    INDEX
    Explanations

    say "ydro/edral word"

    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
    ä¸įåı¯
    -0.77
     immutable
    -0.44
     invariant
    -0.43
     inv
    -0.40
     intrinsic
    -0.39
     ir
    -0.39
     irreversible
    -0.39
     irre
    -0.37
     inex
    -0.36
    ä¸įåıĺ
    -0.36
    POSITIVE LOGITS
    骷
    0.27
    ogy
    0.26
    å¾·æĭī
    0.25
    é²ĩ
    0.24
     dropped
    0.24
    offsetof
    0.23
    еÑħ
    0.23
    xes
    0.23
     drop
    0.23
    .bs
    0.23
    Activations Density 6.898%

    No Known Activations