© 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. 33-TRANSCODER-HP
    7. 54766
    Prev
    Next
    INDEX
    Explanations

    without/with

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    å®īå®ģ
    -0.27
    第ä¸ĥ
    -0.27
    emed
    -0.27
    稷
    -0.26
     Fourth
    -0.26
    æĺİçŁ¥
    -0.26
    èĢĮè¨Ģ
    -0.25
    让ä»ĸ们
    -0.24
    '].$
    -0.24
    ä¸Ģå¹´ä¸Ģ度
    -0.24
    POSITIVE LOGITS
    ric
    0.30
     desire
    0.29
     detail
    0.29
     consequence
    0.29
     merit
    0.29
     cost
    0.28
    presso
    0.27
     deep
    0.27
     major
    0.27
     room
    0.27
    Activations Density 0.007%

    No Known Activations