© 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. 43331
    Prev
    Next
    INDEX
    Explanations

    coding/data

    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
    annah
    -0.29
    纤
    -0.29
    ibal
    -0.26
    è°·
    -0.26
     Homework
    -0.25
     timetable
    -0.25
    -face
    -0.24
    æĶ¿åįı
    -0.24
    åķĨ
    -0.24
    ç¼Ł
    -0.24
    POSITIVE LOGITS
    apl
    0.29
     oceans
    0.29
    æ¹ĸæ³Ĭ
    0.28
     lakes
    0.28
     lanes
    0.28
    æĴ¸
    0.27
    rust
    0.27
     streams
    0.27
    éľ²
    0.27
    positories
    0.26
    Activations Density 0.001%

    No Known Activations