© 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. 41657
    Prev
    Next
    INDEX
    Explanations

    technical, computer-related text

    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
     presentation
    -1.26
     presenting
    -1.16
     presented
    -1.15
     presentations
    -1.13
     Presentation
    -1.10
    Presentation
    -1.04
     presents
    -1.04
    presentation
    -1.01
     presenter
    -0.91
     Presented
    -0.90
    POSITIVE LOGITS
    æ¤ħ
    0.29
    luent
    0.27
    кажеÑĤ
    0.26
    èĤ²äºº
    0.26
    rex
    0.26
     bras
    0.25
    rias
    0.24
    saida
    0.24
     før
    0.23
    nesia
    0.23
    Activations Density 0.001%

    No Known Activations