© 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. 109330
    Prev
    Next
    INDEX
    Explanations

    Technical topics

    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
    èĿĹ
    -0.26
     Pew
    -0.24
    limits
    -0.24
     Lisbon
    -0.23
     bufio
    -0.23
    veh
    -0.23
    æľĢéĢĤåIJĪ
    -0.23
    ç¬ĭ
    -0.23
     Psycho
    -0.23
     dup
    -0.23
    POSITIVE LOGITS
    æĪĸ
    0.33
     or
    0.30
    æĪĸèĢħ
    0.29
    ê±°ëĤĺ
    0.29
    å¥
    0.29
    —are
    0.27
    DSL
    0.27
     æĪĸ
    0.26
     atau
    0.25
     или
    0.25
    Activations Density 0.979%

    No Known Activations