© 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. 52561
    Prev
    Next
    INDEX
    Explanations

    code/technical errors

    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
    t
    -0.39
     ar
    -0.28
    tic
    -0.27
    åħĭ
    -0.27
    dat
    -0.26
    des
    -0.25
    amo
    -0.25
    tur
    -0.25
    æł¼
    -0.25
     ou
    -0.24
    POSITIVE LOGITS
    ypad
    0.30
    inka
    0.29
    eltas
    0.26
    oft
    0.26
    enton
    0.25
     uninitialized
    0.25
    fbe
    0.25
    饷
    0.25
    matchCondition
    0.25
    sein
    0.25
    Activations Density 0.007%

    No Known Activations