© 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. 52866
    Prev
    Next
    INDEX
    Explanations

    Language or text-related

    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
    Not in Any Lists

    No Comments

    Negative Logits
    tring
    -0.27
    ãģ¾ãģĦ
    -0.26
    iento
    -0.26
    -plane
    -0.25
    rien
    -0.25
    amic
    -0.25
    bew
    -0.24
    sett
    -0.24
     bri
    -0.24
    ingga
    -0.24
    POSITIVE LOGITS
    æ´ŀ
    0.28
    extended
    0.27
     temper
    0.27
    drops
    0.26
     Stage
    0.26
     broken
    0.25
    овал
    0.25
    [list
    0.24
     technical
    0.23
     ne
    0.23
    Activations Density 0.018%

    No Known Activations