© 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. 17823
    Prev
    Next
    INDEX
    Explanations

    ,

    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
     attendance
    -0.26
    æ´»
    -0.26
    év
    -0.26
    客车
    -0.25
    æĮª
    -0.25
    logue
    -0.24
    æĨİ
    -0.24
     rendition
    -0.23
     palate
    -0.23
     palette
    -0.23
    POSITIVE LOGITS
     belonged
    0.31
    ç»ĵæĿŁäºĨ
    0.30
    .isOn
    0.29
    reo
    0.27
    ayan
    0.26
    çĭ¬ç«ĭ
    0.26
    åıĺäºĨ
    0.26
    ECH
    0.26
    æİīäºĨ
    0.26
    åİ»äºĨ
    0.25
    Activations Density 0.005%

    No Known Activations