© 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. 43499
    Prev
    Next
    INDEX
    Explanations

    in

    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
     line
    -0.36
    ccoli
    -0.31
    	line
    -0.28
    ÑĢак
    -0.28
    -line
    -0.27
     lines
    -0.26
    .Marshal
    -0.26
    _line
    -0.26
    line
    -0.26
    .line
    -0.26
    POSITIVE LOGITS
    æĶ¯æĮģ
    0.34
    空éĹ´
    0.32
     space
    0.31
     sout
    0.30
    æĶ¯æĴij
    0.30
    .support
    0.30
     supporting
    0.29
    -s
    0.28
    -support
    0.27
    Support
    0.27
    Activations Density 0.001%

    No Known Activations