© 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. 109292
    Prev
    Next
    INDEX
    Explanations

    public figures

    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
    uter
    -0.32
    èĥ½åģļåĪ°
    -0.28
    ç¢Ľ
    -0.27
    cheme
    -0.25
    âĢ¦”↵↵
    -0.25
    èĥ½æ»¡è¶³
    -0.25
    å¾Ĺèµ·
    -0.25
    chemes
    -0.25
    ...">↵
    -0.24
    åİĭ缩
    -0.24
    POSITIVE LOGITS
    typeid
    0.29
    è¾ĩ
    0.28
    's
    0.28
    epad
    0.26
    ’s
    0.25
    تص
    0.23
     makeup
    0.23
     Makeup
    0.23
    çļĦç»ıåħ¸
    0.23
     affair
    0.23
    Activations Density 1.002%

    No Known Activations