© 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. 41970
    Prev
    Next
    INDEX
    Explanations

    the

    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
    å°ıé¾Ļ
    -0.28
    land
    -0.27
    .poly
    -0.26
    å´Ĩ
    -0.26
    æ®ī
    -0.26
    ilage
    -0.24
    lien
    -0.24
     Routes
    -0.24
    è°ł
    -0.24
    笺
    -0.23
    POSITIVE LOGITS
     material
    0.30
     Mater
    0.29
    ä¸įä»ħæĺ¯
    0.28
     Material
    0.27
    å¸ĿåĽ½
    0.27
    é»Ħéĩij
    0.26
    æĿIJæĸĻ
    0.26
    Material
    0.25
     flush
    0.25
    material
    0.25
    Activations Density 0.000%

    No Known Activations