© 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. 62659
    Prev
    Next
    INDEX
    Explanations

    Disconnected text snippets

    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
     CONSEQUENTIAL
    -0.28
    nextInt
    -0.26
    ailing
    -0.25
    EXPR
    -0.25
    /library
    -0.25
     sort
    -0.24
    ä¼ĹæīĢ
    -0.24
    éĢģç»Ļ
    -0.24
    è¯ī
    -0.24
     overlapping
    -0.24
    POSITIVE LOGITS
    orest
    0.28
    ffe
    0.25
     Figure
    0.25
     Typed
    0.25
     Tear
    0.25
    æĦļ
    0.25
     gore
    0.25
    ovid
    0.25
     MAT
    0.24
     volunt
    0.24
    Activations Density 16.946%

    No Known Activations