© 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. 130026
    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
    artment
    -0.29
    åŁºç¡Ģ
    -0.27
    leÅŁtir
    -0.25
    å¢ģ
    -0.25
    溱
    -0.25
    icated
    -0.24
    æľĢ
    -0.24
    fäh
    -0.24
    decess
    -0.24
    tern
    -0.24
    POSITIVE LOGITS
    帶ä¾Ĩ
    0.25
     channelId
    0.25
    âĢ³
    0.25
     rum
    0.25
    èĤ¡æľ¬
    0.24
    å«ģç»Ļ
    0.23
     Transcript
    0.23
    worth
    0.23
     :</
    0.22
    ç½²
    0.22
    Activations Density 10.824%

    No Known Activations

    This feature has no known activations.