© 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. 108174
    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
    èĤ¤
    -0.27
     motivating
    -0.27
    .perform
    -0.25
     minds
    -0.25
     contracting
    -0.25
    owl
    -0.24
    bang
    -0.24
     motiv
    -0.24
    inite
    -0.24
    ç£ķ
    -0.24
    POSITIVE LOGITS
    å¾Ĺ好
    0.29
    âī¦
    0.27
    True
    0.26
    ata
    0.26
    ably
    0.25
    False
    0.25
    uyá»ĥn
    0.25
     UnityEngine
    0.24
    So
    0.24
    odafone
    0.24
    Activations Density 8.560%

    No Known Activations

    This feature has no known activations.