© 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. 20-TRANSCODER-HP
    7. 62948
    Prev
    Next
    INDEX
    Explanations

    Punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lage
    -0.30
    -io
    -0.29
    åħ¬è¯ī
    -0.28
    Located
    -0.27
     centres
    -0.26
    ç»ıåİĨäºĨ
    -0.26
    éħ¿éĢł
    -0.26
    å¤įæĿĤçļĦ
    -0.26
    åı¤èĢģçļĦ
    -0.26
    zs
    -0.25
    POSITIVE LOGITS
    éĢļ常æĺ¯
    0.30
    åij¨åĪĬ
    0.29
    ız
    0.25
    çĭ¬
    0.25
     breath
    0.25
    ä¾Ľ
    0.24
    uteur
    0.24
    umph
    0.24
    ãģ£ãģ½
    0.24
    èĥ¥
    0.24
    Activations Density 9.027%

    No Known Activations