© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 62705
    Prev
    Next
    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash
    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
    Negative Logits
     likewise
    -0.35
    ä¹Łæĺ¯å¦ĤæѤ
    -0.33
     Likewise
    -0.31
    obile
    -0.28
    eded
    -0.28
    ä¹Łä¸įä¾ĭå¤ĸ
    -0.28
     similarly
    -0.27
    Same
    -0.27
    è·ĥ
    -0.25
    ouses
    -0.25
    POSITIVE LOGITS
    Ops
    0.27
    èĪģ
    0.27
    _ops
    0.26
    icts
    0.26
    tight
    0.25
    anye
    0.25
    çļĦéľĢæ±Ĥ
    0.25
     Inv
    0.24
     EVENTS
    0.24
    ynchronize
    0.24
    Activations Density 7.801%

    No Known Activations