© 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. 19982
    Prev
    Next
    INDEX
    Explanations

    £ symbol

    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
    sonian
    -0.30
    usto
    -0.27
     Cabin
    -0.26
    åī¯åİ¿éķ¿
    -0.26
    cki
    -0.26
     drums
    -0.26
     undis
    -0.25
     ÐłÐµÑģпÑĥб
    -0.25
    Republic
    -0.25
    åİ¿å§Ķ书记
    -0.25
    POSITIVE LOGITS
    s
    0.33
    æģ¤
    0.27
     uniform
    0.26
    èĪĮ
    0.26
    ungs
    0.26
    èĮİ
    0.26
    ToStr
    0.25
    åĨľæĿij
    0.25
    ĥĿ
    0.24
    ãĥ¼ãĥĪ
    0.24
    Activations Density 0.026%

    No Known Activations