© 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. 30-TRANSCODER-HP
    7. 96950
    Prev
    Next
    INDEX
    Explanations

    technical and scientific articles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Middle
    -0.36
    Middle
    -0.32
    udos
    -0.30
    _middle
    -0.29
     moyen
    -0.28
    .middle
    -0.27
     middleware
    -0.27
    æ¥Ķ
    -0.25
    _MIDDLE
    -0.25
     medio
    -0.25
    POSITIVE LOGITS
    éĹ»
    0.28
    ITES
    0.28
    -feed
    0.27
    cline
    0.26
    LOPT
    0.24
     SW
    0.24
     SMART
    0.24
    åIJ¬è¯´
    0.24
    SW
    0.23
    è¾ħ
    0.23
    Activations Density 0.932%

    No Known Activations