© 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. 56679
    Prev
    Next
    INDEX
    Explanations

    alo

    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
    Not in Any Lists

    No Comments

    Negative Logits
     sift
    -0.29
     tán
    -0.27
    _DF
    -0.26
    ecret
    -0.24
    casts
    -0.24
     relations
    -0.24
     Directions
    -0.24
    è¡¥
    -0.24
    ybrid
    -0.23
     conventional
    -0.23
    POSITIVE LOGITS
    å·²ä¹ħ
    0.29
    meet
    0.26
    vara
    0.26
    峡
    0.25
    è´
    0.25
     rozwiÄħzania
    0.24
    代çIJĨ
    0.24
     ComVisible
    0.23
     preceded
    0.23
    Ef
    0.23
    Activations Density 0.002%

    No Known Activations