© 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. 27-TRANSCODER-HP
    7. 58643
    Prev
    Next
    INDEX
    Explanations

    code snippets

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     slump
    -0.29
    /story
    -0.26
    è¿ĩå¾Ĺ
    -0.26
     Explos
    -0.25
    olph
    -0.24
     Cong
    -0.24
    fällig
    -0.24
    æµģåĬ¨
    -0.24
    æī¼
    -0.24
     grounded
    -0.23
    POSITIVE LOGITS
    -confidence
    0.29
    eras
    0.29
    ä¸ĢéĿ¢
    0.28
    çĶ¨éĩı
    0.26
     fiat
    0.26
     confession
    0.26
    .requests
    0.26
     retros
    0.25
    çĿģ
    0.25
    vens
    0.24
    Activations Density 0.013%

    No Known Activations