© 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. 33-TRANSCODER-HP
    7. 52797
    Prev
    Next
    INDEX
    Explanations

    find

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_33.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.33.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Manit
    -0.29
    æĹ¥è®¯
    -0.27
     ecc
    -0.26
    omial
    -0.25
     qualifiers
    -0.25
    FFE
    -0.25
    waters
    -0.25
    ثاÙĦ
    -0.25
    èį¼
    -0.24
     Marvin
    -0.24
    POSITIVE LOGITS
    åĩº
    0.54
     out
    0.43
    out
    0.36
    åĩºäºĨ
    0.35
    ,out
    0.34
    -out
    0.33
     Out
    0.33
    =out
    0.32
    	out
    0.32
    Out
    0.32
    Activations Density 0.011%

    No Known Activations