© 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. 129063
    Prev
    Next
    INDEX
    Explanations

    academic publications

    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
     accommodation
    -0.28
    sehen
    -0.27
    åľ°ä¸ĭæ°´
    -0.26
    asts
    -0.25
    .Handled
    -0.25
    на
    -0.25
    uggle
    -0.25
    [thread
    -0.24
    _handling
    -0.24
    æĸ¼æĺ¯
    -0.24
    POSITIVE LOGITS
     vrij
    0.25
    authorize
    0.25
    éŀŃ
    0.24
    trim
    0.24
    ICTURE
    0.24
    积æŀģåĵįåºĶ
    0.24
    compressed
    0.23
    å§Ķåijĺ
    0.23
    æĹłä¸į
    0.23
    weet
    0.23
    Activations Density 0.506%

    No Known Activations