© 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. 142142
    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
     outweigh
    -0.26
     redhead
    -0.26
    è¿Ļåľº
    -0.24
    arked
    -0.24
    #$
    -0.23
    Å·
    -0.23
    ">$
    -0.23
    XT
    -0.23
    ãģķãģĽãĤĭ
    -0.23
     Rex
    -0.22
    POSITIVE LOGITS
    agrams
    0.28
    isme
    0.25
    代
    0.24
    .fromString
    0.23
    /us
    0.23
    IZER
    0.23
    MBER
    0.23
    æľī两ç§į
    0.23
     Marvel
    0.23
     Drawing
    0.23
    Activations Density 8.968%

    No Known Activations

    This feature has no known activations.