© 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. 20-TRANSCODER-HP
    7. 64072
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    annel
    -0.27
    à¹Ģà¸Ńà¸ĩ
    -0.26
    cil
    -0.26
    äºĨèĩªå·±çļĦ
    -0.26
    æľīä¿¡å¿ĥ
    -0.25
    rowable
    -0.25
    èħ±
    -0.25
     mình
    -0.25
    èĩªå·±
    -0.24
    uran
    -0.23
    POSITIVE LOGITS
     balance
    0.27
    hold
    0.27
    æĸ¯å¡Ķ
    0.26
    -bal
    0.25
    oyo
    0.25
    holds
    0.24
    fell
    0.24
     stimulation
    0.24
    -weight
    0.24
     creation
    0.24
    Activations Density 5.196%

    No Known Activations

    This feature has no known activations.