© 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. 108113
    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
    els
    -0.29
    ices
    -0.27
     sm
    -0.27
    -di
    -0.26
    sm
    -0.26
    失èIJ½
    -0.25
    (sm
    -0.25
    åįī
    -0.25
    çľ¼åīį
    -0.25
    -sm
    -0.25
    POSITIVE LOGITS
    è¾¾æĪIJ
    0.26
    æĥ³å¿µ
    0.25
    ç«¿
    0.25
    åij¨æģ©
    0.24
     Diá»ħn
    0.24
    éŃĶ鬼
    0.24
    iad
    0.24
    ç½Ķ
    0.24
    imonial
    0.24
    rame
    0.23
    Activations Density 16.849%

    No Known Activations

    This feature has no known activations.