© 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. 43208
    Prev
    Next
    INDEX
    Explanations

    uniqueness and distinction

    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
    z
    -0.28
    lood
    -0.28
    ad
    -0.27
    ä»»ä½ķ
    -0.27
    ady
    -0.27
     anyone
    -0.25
     craz
    -0.25
    arp
    -0.25
    osen
    -0.25
    (any
    -0.24
    POSITIVE LOGITS
     distinct
    0.89
    distinct
    0.73
     unique
    0.64
     DISTINCT
    0.62
    Distinct
    0.61
     distinctive
    0.56
     uniquely
    0.53
     Unique
    0.50
    unique
    0.48
     distinctly
    0.47
    Activations Density 0.002%

    No Known Activations