© 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. 27-TRANSCODER-HP
    7. 9692
    Prev
    Next
    INDEX
    Explanations

    unstructured text

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_27.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.27.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    éĽı
    -0.28
     fondo
    -0.26
    (thing
    -0.25
    家
    -0.25
    angen
    -0.25
    سب
    -0.24
     spanning
    -0.24
     Neg
    -0.24
    ä»Ģä¹Īåij¢
    -0.23
    ä»Ģä¹Ī
    -0.23
    POSITIVE LOGITS
    pak
    0.26
    cripts
    0.25
    ulous
    0.25
    á»ĩn
    0.25
    ettes
    0.25
     ServiceException
    0.24
    ffd
    0.24
    soles
    0.24
     dissip
    0.24
     flour
    0.24
    Activations Density 0.127%

    No Known Activations