© 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. 17813
    Prev
    Next
    INDEX
    Explanations

    financial, code-related texts

    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
     lie
    -0.31
     lies
    -0.28
     Contr
    -0.27
     li
    -0.26
     credit
    -0.26
    clerosis
    -0.26
    omba
    -0.25
    èħ¿
    -0.25
     tack
    -0.25
    <li
    -0.25
    POSITIVE LOGITS
    亲è¿ij
    0.31
     mij
    0.28
    å®ĩå®Ļ
    0.26
    gü
    0.26
    spir
    0.26
     Transit
    0.26
    inspace
    0.26
    è¿·
    0.25
    NL
    0.25
    æ´»
    0.25
    Activations Density 0.213%

    No Known Activations