© 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. 62539
    Prev
    Next
    INDEX
    Explanations

    articles

    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
    mates
    -0.26
     Pew
    -0.26
     mates
    -0.25
    æ¾Ħæ¸ħ
    -0.25
    ción
    -0.24
     dare
    -0.23
     eternity
    -0.23
     Eaton
    -0.23
    -camera
    -0.23
    ghan
    -0.23
    POSITIVE LOGITS
    åľ¨è¿Ļ
    0.28
    reference
    0.26
    ä»İä¸ļ
    0.25
    æī¾åĪ°
    0.25
    -LAST
    0.25
    >:</
    0.25
    æŁ¥æī¾
    0.24
    PI
    0.24
    çĭ¬
    0.24
    pons
    0.24
    Activations Density 16.889%

    No Known Activations