© 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. 54596
    Prev
    Next
    INDEX
    Explanations

    text snippets

    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
     ton
    -0.36
    ton
    -0.29
    avor
    -0.27
     Meth
    -0.26
    utor
    -0.26
    eday
    -0.26
    út
    -0.26
    utron
    -0.26
    dit
    -0.25
    ÑĥÑĤÑĮ
    -0.25
    POSITIVE LOGITS
     lump
    0.34
    åĮ¯
    0.27
    ioc
    0.26
    èĭ¹æŀľ
    0.26
     banana
    0.26
    ilated
    0.26
     Holy
    0.25
     Rash
    0.25
    camp
    0.25
     ViewGroup
    0.25
    Activations Density 0.270%

    No Known Activations