© 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. 109307
    Prev
    Next
    INDEX
    Explanations

    Contains two dates

    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
    æ¸ħæĺİ
    -0.28
     respects
    -0.28
    ettes
    -0.28
    å°Ĭéĩį
    -0.27
     respect
    -0.26
    Discover
    -0.26
    äºĮåįģåĽĽ
    -0.26
    agine
    -0.25
    çĩĥæĸĻ
    -0.25
    梵
    -0.25
    POSITIVE LOGITS
    ldb
    0.28
    idor
    0.28
    bor
    0.28
    tones
    0.27
    é«ĺæ°´å¹³
    0.26
    prov
    0.26
    PEnd
    0.24
    cke
    0.24
    _tb
    0.24
     Bloom
    0.24
    Activations Density 0.140%

    No Known Activations