© 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. 43175
    Prev
    Next
    INDEX
    Explanations

    genetics and mutations

    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
    utr
    -0.27
    paper
    -0.26
    hoe
    -0.25
    redd
    -0.24
     passes
    -0.24
    å½·
    -0.24
    èĢĻ
    -0.24
     Doub
    -0.24
     nær
    -0.24
    ÑĮ
    -0.23
    POSITIVE LOGITS
     Inf
    0.30
    ines
    0.29
     inf
    0.29
    ä¸ģ
    0.28
     conduct
    0.28
    诱导
    0.27
    Inf
    0.27
     indifference
    0.27
    CAP
    0.27
     dine
    0.27
    Activations Density 0.011%

    No Known Activations