© 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. 25-TRANSCODER-HP
    7. 106950
    Prev
    Next
    INDEX
    Explanations

    biological/technical responses

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     tabindex
    -0.29
     tease
    -0.27
    ainen
    -0.26
    ëĵĿ
    -0.26
    {{{
    -0.25
    建ç«ĭäºĨ
    -0.25
    ativo
    -0.25
    ioned
    -0.25
    建ç«ĭ
    -0.24
    第ä¸ī次
    -0.24
    POSITIVE LOGITS
    åŁ½
    0.29
    cur
    0.29
     Cur
    0.28
    åľ°éģĵ
    0.28
     cur
    0.28
     CUR
    0.27
    CUR
    0.27
    管å±Ģ
    0.27
     curs
    0.26
    Cur
    0.26
    Activations Density 0.910%

    No Known Activations