© 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. 32-TRANSCODER-HP
    7. 41888
    Prev
    Next
    INDEX
    Explanations

    mathematical notation

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    å½ĵåľº
    -0.27
    *(-
    -0.26
     vacancies
    -0.26
    /lgpl
    -0.25
    SITE
    -0.25
     capitals
    -0.25
    <=(
    -0.25
     °
    -0.25
     *,
    -0.25
     Gund
    -0.24
    POSITIVE LOGITS
    ^
    0.42
    ^\
    0.37
    ^{
    0.37
    '^
    0.36
    ^-
    0.33
    ^n
    0.33
    ^{-
    0.29
    æĬĬä½ł
    0.29
    ingly
    0.28
    éħį
    0.28
    Activations Density 0.157%

    No Known Activations