© 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. 42326
    Prev
    Next
    INDEX
    Explanations

    Non-English words

    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
    rates
    -0.29
    shaw
    -0.28
    rm
    -0.27
    åģı
    -0.26
    rug
    -0.26
    ersh
    -0.26
    meter
    -0.25
    cript
    -0.25
    &m
    -0.25
    æ¡Ģ
    -0.25
    POSITIVE LOGITS
    ä¹°åįĸ
    0.27
     Franç
    0.25
     flick
    0.24
    \Tests
    0.24
    nce
    0.24
    å°ıå§IJå§IJ
    0.23
     Beat
    0.23
    è°ĥåīĤ
    0.22
    å¯¹ä½ł
    0.22
     seiz
    0.22
    Activations Density 0.002%

    No Known Activations