© 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. 62728
    Prev
    Next
    INDEX
    Explanations

    writing

    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
    Negative Logits
    éĥ¨
    -0.29
    .ravel
    -0.27
     Ill
    -0.25
    oulder
    -0.25
    顧
    -0.25
    客
    -0.25
     atomic
    -0.25
    :-
    -0.24
    ,
    -0.24
     SHA
    -0.24
    POSITIVE LOGITS
    gles
    0.31
    uele
    0.28
    ç«ĭå¿Ĺ
    0.28
    IELDS
    0.27
     Spare
    0.27
    stin
    0.26
    statt
    0.26
    ç»ĻåĩºäºĨ
    0.26
     discovers
    0.25
     hurry
    0.25
    Activations Density 1.891%

    No Known Activations