© 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. 63244
    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
    ampie
    -0.30
    pitch
    -0.28
     pitches
    -0.28
     (++
    -0.26
     Word
    -0.26
    çľ¼æ³ª
    -0.26
    orrent
    -0.25
     Semantic
    -0.25
    igger
    -0.24
    Verb
    -0.24
    POSITIVE LOGITS
    ëĿ½
    0.28
    åŃľ
    0.27
    èµ°åIJij
    0.27
    éªŀ
    0.27
    è¿ŀ线
    0.26
    illon
    0.25
    hcp
    0.24
    æĦı大
    0.24
     Fir
    0.24
     quir
    0.23
    Activations Density 59.142%

    No Known Activations

    This feature has no known activations.