© 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. 129734
    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
    ç§ĭåĨ¬
    -0.27
    +p
    -0.26
    aveled
    -0.25
     pir
    -0.24
    orra
    -0.24
    å¼Ģå§ĭäºĨ
    -0.24
    è¡ħ
    -0.23
    éĥ½ä¼ļæľī
    -0.23
    榴
    -0.23
    ä¸Ģ线
    -0.23
    POSITIVE LOGITS
     disc
    0.27
     Chand
    0.26
    ç»ı
    0.26
    ç§įç±»
    0.25
     pasado
    0.25
    chan
    0.25
    æĭĸ
    0.25
    agua
    0.24
    æºŁ
    0.24
     crit
    0.24
    Activations Density 19.347%

    No Known Activations

    This feature has no known activations.