© 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. 31-TRANSCODER-HP
    7. 17969
    Prev
    Next
    INDEX
    Explanations

    Code snippets/technical documents

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     CommandType
    -0.28
    YST
    -0.27
     Cupertino
    -0.27
    ;height
    -0.26
    cycl
    -0.26
     Gibson
    -0.25
    cycles
    -0.25
    éĻ¢
    -0.25
    _sb
    -0.24
    æ·±åĪĩ
    -0.24
    POSITIVE LOGITS
     med
    0.35
     def
    0.28
    è¦ģ害
    0.27
    æĴŃç§į
    0.27
    woke
    0.26
    åĪ³
    0.26
    æī³
    0.26
    æĹħ客
    0.25
    ç³Ĭ
    0.25
    ник
    0.24
    Activations Density 1.082%

    No Known Activations